University of Southern California Dissertations and Theses

Change Strategy And Client System: Administrative Reform In Brazil

(USC Thesis Other)

Change Strategy And Client System: Administrative Reform In Brazil

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been microfilmed exactly as received 6 6—7078 NASCIMENTO, Kleber Tatinge D o , 1 9 3 7 - CHANGE STRATEGY AND C L IE N T SYSTEM: a d m i n i s t r a t i v e r e f o r m i n b r a z i l . U n iv e r sity o f Southern C aliforn ia., B h .D ., 1966 P o litic a l S c ien ce , public a d m in is tr a tio n U niversity Microfilms, Inc., A nn A r b o r , Michigan Copyright by Kleber Tatinge do Nascimento 1966 CHANGE STRATEGY AND CLIENT SYSTEM: ADMINISTRATIVE REFORM IN BRAZIL by Kleber Tatinge do Nascimento A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Public Administration) June 1956 UNIVERSITY O F SO U T H E R N CALIFORNIA TH E GRADUATE SC HOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 9 0 0 0 7 This dissertation, written by KLEBER TATINGE DO NASCIMENTO under the direction of h ^ . —Dissertation Com mittee, and approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R O F P H I L O S O P H Y .... Dean D a t e . . . l m ................ DISSERTATION COMMITTEE AOS MEUS TRES ANJOS: FARIDE, DAISY, E CLAUDIA. ACKNOWLEDGEMENT In composing these final words of appreciation, I am acutely conscious that no matter what I write it will be nowhere near adequate expression of my gratitude. Actually, I have rewritten this part more times than any other in my dissertation. After several unsatisfactory trials, I feel that the more I write about the competence, understanding, friendship, and devotion of an unforgetable group of people, the more unfair I become as to the actual dimension of their assistance. May I then simply list their names without distorting my gratitude with ever inaccurate comments. Professors Frank P. Sherwood, Alexander W. McEachem, Richard W. Gable, and Gilbert B. Siegel. jMrs. Muriel E. Sipe, Mr. Robert P. Biller, and Dr. Dorothy V. Meredith. Professors Diogo Lordello de Mello, Benedito Britto, Carlos Verfsimo de Almeida Amaral, and Edi Madalena Fracasso. Mr. Jerry R. Marlatt, and Misses Janna L. Gadden and Angelina Aguilar. Messrs. Theodore H. Thomas and John Good, and Mrs. Connie Rodgers. And my wife, Daisy Nascimento. iii TABLE OF CONTENTS Page DEDICATION . ..................... ii ACKNOWLEDGEMENT . iii LIST OF TABLES............................... vii LIST OF FIGURES.............................. x Chapter I. INTRODUCTION ........................... 1 Nature, Aims, and Scope of the Study Rationale of the Study; The Relationship Between Strategy and Obstacles in the Change Process Major Analytical Concepts: Towards Models of Administrative Change Strategy The Research Design Contents of the Dissertation II. TOWARDS MODELS OF ADMINISTRATIVE CHANGE STRATEGY; CONTRIBUTIONS OF THE LITERATURE ON PLANNED CHANGE.................... 49 Levels of Change: From Individual to International Systems Planned Organizational Change: Towards Model Building III. THE BRAZILIAN ADMINISTRATIVE REFORMS AND THEIR STRATEGIC MODEL.................. 126 The Purchasing and Materiel Reforms The Position Classification Reforms iv Chapter v Page The Merit System Reforms The Budgetary Reforms The Structural-Functional Reforms Conclusion IV. AUTHORITY PATTERNS AND THEIR IMPLICATIONS FOR CHANGE STRATEGY: SOME THEORETICAL PROPOSITIONS ....................... Authority Relationships: The Need for Behavioral Approaches Authority Patterns and Planned Organiza tional Change Position-Orientedness in Authority Relationships Authority Relationships: Indicators of Position-Orientedness Functionality and Dysfunctionality of Position-Orientedness in Organizational Behavior Authority Relationships and Comparative Public Administration V. AN ASPECT OF THE BEHAVIORAL PROFILE OF THE BRAZILIAN BUREAUCRACY: AUTHORITY PATTERNS ........................... An Empirical Study: Rationale Status-Protectiveness: Findings Emphasis on Hierarchy: Findings Extension of the Scope of the Hierarchi cal Relationship: Findings Findings of the Study: Over-all Interpretation Summary and Conclusions 228 274 VI. ANOTHER ASPECT OF THE BEHAVIORAL PROFILE OF THE BRAZILIAN BUREAUCRACY: TEMPORAL ORIENTATIONS..................... 336 Chapter Time and Human Behavior: A Conditioning Relationship The Instrumental Model and the Temporal Prerequisites of Its Strategy Ideology Temporal Orientations in the Brazilian Bureaucracy: An Empirical Study Findings of the Study Summary and Conclusions VII. CONCLUSIONS ........................... Summary An Interpretation of the Past A Proposal for the Future BIBLIOGRAPHY .............................. APPENDIX A: AUTHORITY PATTERNS QUESTIONNAIRE . . . APPENDIX B: AUTHORITY PATTERNS— CODING CATEGORIES. APPENDIX C: TIME PATTERNS QUESTIONNAIRE ....... APPENDIX D: TIME PATTERNS— CODING CATEGORIES . . . vi Page 417 447 464 481 489 497 LIST OF TABLES Table 1. 2. 3. 4. 5. 6. 7. Assumptions and Approaches of Three Analytic Models ...................... A Comparison Between Chin1s Models and the Instrumental and Noninstrumental Models of Change Strategy .................... Proportion of Positions Subject to Competi tion in the Federal Service, 1884-1960 . ■ . Position-Orientedness as Indicated by Status- Protectiveness (Items 1 to 5, Part II) Proportions of Responses to the Question "What Actually Happens?" and Response Categories Used in Parentheses ......... Position-Orientedness as Indicated by Status- Protectiveness (Items 1 to 5, Part II) Proportions of Responses to the Question "What Should Happen?" and Response Categories used in Parentheses ......... Position-Orientedness as Indicated by Emphasis on Hierarchy (Items 6 to 19, Part II) Proportions of Responses to the Question "What Actually Happens?" and Response Categories Used in Parentheses. . Position-Orientedness as Indicated by Emphasis on Hierarchy (Items 6 to 19, Part II) Proportions of Responses to the Question "What Should Happen?" and Response Categories Used in Parentheses. . Page 78 82 174 298 300 305 309 vii viii Table Page S. Position-Orientedness as Indicated by Exten sion of the Scope of the Hierarchical Relationship (Items 20 to 24, Part II) Proportions of Responses to the Question "What Actually Happens?" and Response Categories Used in Parentheses......... 313 9. Position-Orientedness as Indicated by Exten sion of the Scope of the Hierarchical Relationship (Items 20 to 24, Part II) Proportions of Responses to the Question "What Should Happen?" and Response Categories Used in Parentheses......... 316 10. A g e ................................... 373 11. Supervision Experience ................... 373 12. Length of Supervision Experience ...... 374 13. Definition of Lateness................... 377 14. Tolerance of Lateness................... 373 15. Reaction to Lateness (Social Matters) ... 379 16. Prescribed Working Load................. 381 17. Respondents' Felt Sense of Urgency........ 382 18. Governmental Personnel's Work Pace........ 383 19. Time Inputs of Top Governmental Personnel . 384 20. Relative Time Inputs Between Public and Private Personnel (Higher Echelons) . . . 386 21. Relative Time Inputs Between Public and Private Personnel (Lower Echelons) .... 386 ix Table Page 22. Scheduling Habit (Appointments) ...... 389 23. Reaction to Lack of Scheduling (Appoint ments ............................... 391 24. Scheduling Habit (Typical Work Day— Perceptions) ..... ................ 392 25. Scheduling Habit (Typical Work Day— Expectations) ....................... 392 26. Respondents' Time Perspective........... 395 27. Direction of Time Perspective in Governmental Decision-Making...................... 399 28. Governmental Temporal Orientations— Perceptions ........... ........ 401 29. Governmental Temporal Orientations— Expectations......................... 401 30. Respondents' Jobs' Temporal Orientations— Perceptions......................... 403 31. Respondents' Jobs' Temporal Orientations— Expectations ..... ................ 403 32. Job-Unrelated Allocation of Time on the J o b ................................ 408 33. Dual Employment in Government (Higher Echelons)......... 409 34. Time Inputs Affected by Dual Employment in Government (Higher Echelons) ........... 410 35. Dual Employment in Government (Lower Echelons).................. 412 36. Time Inputs Affected by Dual Employment In Government (Lower Echelons) ........... 412 ix Table Page 22. Scheduling Habit (Appointments) ......... 389 23. Reaction to Lack of Scheduling (Appoint ments .............................. 391 24. Scheduling Habit (Typical Work Day— Perceptions) ...... 392 25. Scheduling Habit (Typical Work Day— Expectations) ....................... 392 26. Respondents1 Time Perspective........... 395 27. Direction of Time Perspective in Governmental Decision-Making................ 399 28. Governmental Temporal Orientations-- Perceptions......................... 401 29. Governmental Temporal Orientations— Expectations .................. 401 30. Respondents' Jobs' Temporal Orientations— Perceptions.......... 403 31. Respondents' Jobs' Temporal Orientations— Expectations ........... ........ 403 32. Job-Unrelated Allocation of Time on the Job .......... 408 33. Dual Employment in Government (Higher Echelons) ..... .................. 409 34. Time Inputs Affected by Dual Employment in Government (Higher Echelons) . ....... 410 35. Dual Employment in Government (Lower Echelons) .................. ..... 412 36. Time Inputs Affected by Dual Employment in Government (Lower Echelons) ....... 412 LIST OF FIGURES Figure Page 1. Classification of Obstacles to Change ... 17 2. The Instrumental Model of Change Strategy. . 26 3. A Noninstrumental Model of Change Strategy . 29 4. The Two Models of Change Strategy ..... 52 5. Success Probabilities of Intended Change, in Terms of the Strategy Ideology of the Change Model and the Behavioral Profile of the Client System.................. 236 6. An Outline of the Weberian Ideal-Type , . . 259 7. Contradictory Functional Requirements of Bureaucracy ..... ................ 262 x CHAPTER X INTRODUCTION The second half of the forties witnessed a pheaome- nal growth of a comparative approach to social sciences. The postwar reconstruction effort, the closer interdepend- ence between developed and developing nations, the cold war, the emergence of foreign aid and technical assistance as the central goals of various national and supranational entities— these are some of the factors which have deepened the perspectives of social science, by making it less parochial. The fact that both new and old developing nations depend largely— sometimes almost solely— on governmental action for most collective accomplishments suggests why public administration and political science have been among the most deeply revolutionized fields of social inquiry in the last twenty years. The leitipotiv of the comparative approach in pub lic administration, since the post-World War XI period,, 1 has been change, deliberately induced change. It was necessary to modernize anachronistic structures, to en hance the implementative ability of governments, to reformulate goals, to inculcate values, to secularize patterns of thought— it was imperative to change. The inducement of change is an old concern of sciences and, indeed, one of the goals of science. From psychology to anthropology, from mind to culture, the com pulsion to effect change--intended and controlled change— certainly stands among the strongest motivations of social scientists. But only recently has the concern for induced change been conceptualized in an Integrated set of proposi tions. To be sure, planned change seems to be well on its way to becoming a social discipline.-^ This is a comparative study on planned administra tive change in Brazil. Though it tends to be more — ^fc- ... — — r n ■ «H ^ I n' »fci* 1-The two most integrative books dealing explicitly with planned change are: Ronald Lippitt, Jeanne Watson, and Bruce Westley, The Dynamics, of Planned Change (New York: Harcourt, Brace & World, Inc., 1958); and Warren G. Bennis, Kenneth 0. Benne, and Robert Chin (eds.), The planning of Change (New York: Holt, Rinehart & Winston, Inc., 1961). A review of the pertinent literature is pro vided in Chapter II. idiographic than nomothetic,^ the study is "comparative" in at least three senses. First, several of its basic analytical tools are either originated or vastly applied in comparative administration studies. Second, the exper ience of other countries, particularly that of the United States, is often resorted to as a referent against which the Brazilian case is contrasted. Finally, this work may serve as a potential source of methodology for future com parative endeavors on both the theory and practice of planned organizational change, especially in connection with public administration reforms in developing nations. Because it centers on strategies and tactics which oriented administrative reforms, this is also a study in planned change. Indeed, the chief concern of this work is to explore the implications of a given change strategy on the outcomes of administrative reform attempts in Brazil. 2Fred W. Riggs, "Trends in the Comparative Study of Public Administration," jgtrative Sciences. XXVII, No. 1 (1962), 9-15. 4 NatureT Afo«. and Scope of the Study This is a descriptive study aimed at identifying the degree of compatibility between two sets of variables— those required by the strategic pattern of the proposed administrative reforms, on the one hand, and those which prevail in the Brazilian bureaucracy, on the other. Though mainly descriptive in nature, this work has two secondary characteristics. It is historical to the extent that it accounts for the successive trials in reforming public administration in Brazil. The modem history of administrative reform, which began in the thirties under Getulio Vargas, is reviewed and interpreted to identify the prevailing pattern of reform strategy. Finally, this study presumes to make a theoretical contribution by introducing some conceptual distinctions which may be valuable to model building in planned change. The specific aims of this study are (1) to suggest an explanation for the failures of administrative reform attempts in Brazil, and (2) to indicate relevant points for the devising of a strategy of administrative change for that country. In order to meet those objectives, two subgoals are envisaged, namely: (1) the identification and definition of the prevailing strategy of administra tive reform; and (2) the analysis of the behavioral pat terns that pervade the Brazilian bureaucracy, as well as of the implications of such patterns to the prevailing strategy of reform. Besides its specific aims, this study has other long-term objectives, the most significant of which are (1) to present a methodological approach to the analysis of administrative reform; (2) to contribute to the Brazilian literature of public administration; (3) to stimulate the interest of Brazilian students in behaviorally-oriented research; (4) to provide a succinct historical survey of Brazil's efforts to improve its federal administration; and (5) to contribute to the gen eral literature on administrative change in developing nations, particularly where that change involves the transference of both "techniques" and "technics"^ from one cultural setting to another. ^These terms are suggested by Harold Lasswell and Abraham Kaplan, Power and Society (New Haven: Yale University Press, 1950), p. 51, for distinguishing between the value-ridden ("techniques") and value-free ("technics") 6 The scope of this study can be defined by several boundaries. On the time dimension, the period analyzed ranges from 1930, when Getulio Vargas first came to power, to March 31, 1964. The year 1930 is a turning point in the administrative history of Brazil. Before that date, there was no conscious and deliberate movement for administrative improvement in the federal government. The quest for efficiency in government gained impetus under Vargas, both because the dictator was always con cerned with the administrative machinery and because his own political survival at that moment depended, to a great extent, upon an efficient bureaucracy. The period of study ends with March 31, 1964, when President Goulart was overthrown by a civilian-military coalition. The restric tions on democratic liberties, the purge of Congress, the postponing of elections, the social turbulence, and the uncertainties of the new regime brought about a completely aspects of a technology. The distinction is applied by Riggs in his contention that bureaucratic changes are likely to occur more rapidly than changes in politics. Fred W. Riggs, "Bureaucrats and Political Development: A Paradoxical View," in Bureaucracy and Political Develop ment. ed. Joseph La Palombara (Princeton, N.J.: Princeton University Press, 1963), pp. 120-67. different environment for administrative reform. This new phase still lacks sufficient historical perspective to warrant well-founded generalisations. Therefore, only en-passant will references be made to events after the April Revolution. On the spatial dimension, this work is limited to the civilian agencies of the "direct administration" of the Executive Branch of the Brazilian federal government. All military organizations are excluded. "Direct admini stration" excludes all autarchic entities^ and public enterprises, although, formally or not, they are part of the Executive Branch. All state and local administrative agencies are equally outside the scope of this analysis. The experience of the United States is often brought to bear. However, as said before, this study tends to be more idiographic than nomothetic; many comparisons are drawn to corroborate viewpoints and assumptions, but the study could never be properly titled, for instance, "Strategies of Administrative Reform: A Comparison Between the Experiences of Brazil and the United States." ^The typical example is the autarauia of social weIfare administration. 8 Within the specified time-space boundaries, it Is necessary to define vhich administrative reforms are to be examined. "Administrative reform" is a loose expression. In its broadest meaning, it encompasses all decisions, even the minor ones, such as the modification of procedures for requesting office supplies in a small organizational unit. However, for the purposes of this research, administrative reforms are those changes which (1) require legislative approval in times of democratic guarantees; and (2) involve alterations of currently established personnel rights such as salary, hierarchical status, and promotion criteria, as well as of formal relationships between public agencies and organs, work methods and flow, and the statutory com petence of organizational units. The requirement of legislative approval is intended to isolate major reforms only. Under Vargas' dictatorship some basic administrative reforms were determined by decree-laws. Such reforms are included in this work, since they would have required legislative action under democra tic conditions. The following reform attempts meet the operational criteria set above and will be examined in this study: (1) the Purchasing and Materiel Reforms, which started in 1931; (2) the Position Classification Reforms, particularly from the Commission of Readjustment of 1936, to the present; (3) the Merit System Reforms, which began in 1934; (4) the Budgetary Reforms, origina ting with the creation of the Budget Commission in the Ministry of Finance in 1939; and (5) the Structural- Functional Reforms which have been proposed since 1952. These reforms were not independent from, but intermingled with, each other. Also, their differentia tion and labeling here are devised by the author for facilitating the analysis alone; there is no general classification of the administrative reforms attempted in Brazil, so that the classification used in this xrork is entirely arbitrary. Rationale of the Study: The Relationship Between Strategy and Obstacles in the Change Process The history of administrative reform in Brazil is already rich enough to warrant an academic attempt to detailed explanation of what these reforms were is provided in Chapter III. 10 identify whether there has been a prevailing pattern of change strategy, as well as the implications of that pattern on the accomplishments of the reforms. A study of the prevailing reform strategies is basic to under standing the fate of past reforms and to increasing the success probabilities of future efforts. A crucial question can be raised at this point: Why not begin by exploring possible resistances to change and proceed with the identification of the strategic pattern afterwards? Why do just the reverse? A decision has been made to search for possible obstacles to change after the strategic pattern of the attempted change is defined, so that those obstacles may be assessed in relation to the strategy adopted and not as absolute difficulties themselves. This is so because, notwithstanding the fact that obstacles do exist which block any change, it is, nevertheless, theoretically defensible and empirically demonstrable that their barring potential varies with different strategies. In other words, certain strategies tend to reinforce and others to weaken the detaining effect of obstacles. The struggle for the merit system in Brazil is an illustration of the 11 way obstacles may be created by a strategy. In Brazil, the merit system has been defined as one in which entrance to the civil service depends upon the candidate's compe tence as tested in competitive examinations. Consequently, the reforms in this area have centered upon selection. If the merit system were redefined as a system in which the civil servant is provided the necessary training to ade quately perform his functions so that his holding of a position be based upon his merit in performing its tasks, the focus of reform would have been training rather than selection. Accordingly, different obstacles might have emerged. Therefore, resistances to change may well be a function of the change strategy; a particular strategy can even evoke resistances that did not antedate the change attempt. A wholistic approach to administrative reform, for example, is likely to engender obstructions that might be dormant under a gradualistic strategy.^ Or a wholistic %!holism and gradualism refer to the degree of inclusiveness of the change attempt. These concepts correspond to Hirschman's comprehensive and piecemeal styles of policy-making, which he also calls ideological and remedial approaches to problem-solving. Albert 0. Hirschman, Journeys Toward Progress (New York: The 12 approach may accomplish more results under totalitarian rather than democratic political conditions. Another point suggests the advantage of assessing obstacles in relation to the strategy adopted, and not as independent variables. A crucial issue in planned change regards the assumption that change is only possible if certain prerequisites are met; or, contrarywise, that change must be strategically devised so as to ignore, or even reject, the idea of preconditions.7 To assume either that a reform depends upon preconditions or that it is a prerequisite itself to other changes bears heavily on the direction, on the pace, on the philosophy, and, conse quently, on the barriers to be encountered by such a Twentieth Century Fund, Inc., 1963), especially p. 248. Though from a somewhat different theoretical perspective, a parallel can also be traced between the wholistic approach and the rational-coraprehensive model of game theory, economics, and statistical decision theory, on the one hand, and between the gradualistic approach and Bray- brooke and Lindblom's disjointed incremental model. David Braybrooke and Charles E. Lindblom, A Strategy of Decision (Mew York: The Free Press of Glencoe, 1963). 7The question of prerequisites to change has been much explored in development economics, particularly in regard to the issue of balanced versus unbalanced economic growth. Hirschman has been a leading figure among those who sponsor an unbalanced strategy of growth. Albert 0. Hirschman, The Strategy of Economic Development (New Haven and London: Yale University Press, 1958). 13 reform. The opposite assumption that there are no pre requisites for reform would likely generate different problems. For all those reasons, it is suggested that the identification of the strategic pattern which has pre vailed in the Brazilian administrative reforms be the first step in the process of understanding past efforts and guiding future projects. Identifying the pattern of reform strategy is use ful insofar as it serves as a reference point in the search for change obstacles. For analytical purposes, a distinction between potential barriers and barriers either reinforced or elicited by the strategy itself might prove helpful. The first type of barriers concerns those characteristics which are deeply ingrained in the culture and are part of the ethos of a society; they might be termed resistances or impediments. The latter barriers are functions of the strategy chosen and might be named incompatibilities.^ Both are obstacles, but their ^The struggle for the merit system in Brazil is again a good illustration of this difference. Though impediments existed, a whole series of incompatibilities were elicited by the strategy itself; should the merit system have been sought through training rather than selection, those incompatibilities would not have been raised, or would have been of a different nature or potency 14 Implications for planned change are quite different. Studies of strategies and tactics can be theoretically enriched and their pragmatic consequences made more pre dictable if a separation between impediments and incom patibilities is found to have analytical properties. The search for obstacles that hinder reform can be further facilitated if such obstacles are categorised on the basis of their nature. In this study, both impediments and incompatibilities will be classified as cognitive, valuational, or systemic. Any sort of informational scarcity represents a cognitive hindrance. In the Brazilian federal government, for instance, one is unable, at the moment, to determine the precise number of public employees, or to estimate the manpower needs for the years ahead. The bureaucracy lacks an accurate data system which would provide such informa tion. Valuational limitations refer to opinions, beliefs, and attitudes that form the individual's Weltanschauung or the organization's character. The belief that the only honest basis for promotion is seniority, for example, is a valuational barrier which a promotion system based on 15 efficiency ratings may not overcome. Finally, systemic obstacles evolve from the problem-solving history of an organization. They refer to habits and rigidities which are usually rooted in history and have become part of the behavioral style of an organi zation. Systemic obstacles do not represent preferences, conscious choices, but rather "reflexive" behaviors trans mitted through the times, because "it has always been this way." A most eloquent example of a systemic limitation to reforms based on "equal pay for equal work" is the remuner ation of revenue personnel in Brazil. The pay system of such personnel comprises two parts: the first is fixed, set at one of the levels of the general pay scale; the other is variable, established on the basis of a percent age over the total tributary revenue. Even without con sidering the inflationary pressures which Brazil has experienced for a long time now, the automatic revenue increases raise such salaries far above the top of the general pay schedules, and often above the remuneration of the president, governor, or mayor, and corresponding leg islative bodies, as the case may be. Though this sort of privilege has now and then aroused some protests, no law 16 has succeeded In changing the practice. The "princes of the Republic1' — as the revenue personnel are often called— are only modem counterparts to the Crown1 s agents who installed the system in the 1530s. The tradition has survived the ages, transmitted from one generation to another, becoming* institutionalized as a systemic charac teristic, independent of personalities or politics. The foregoing classification of obstacles is depicted in Figure 1. The purpose of classifying obstacles as cognitive, valuational, and systemic is not only to facilitate their identification but also to inquire whether any set of obstacles is a more powerful limitation than the others, and, if so, which one. Here lies a rich source of hypoth eses about strategies for administrative change. As it will be shown, the models of change strategy used in this research differ much from each other as to the importance they attribute to cognitive and valuational impediments. To recapitulate, this study attempts to (1) define the prevailing pattern of administrative reform strategies; and (2) identify the main obstacles to that pattern, as a basis for explaining the fate of adminis- Obstacles npedJ Impediments (cultural orientations, obstacles to any Incompatibilities (obstacles elicited or reinforced by a given change strategy) change strategy) Systemic (habits, ways Valuational (attitudes and beliefs Cognitive (information scarcity limitative of the intended change) limitative of the intended change) doing, crystallized behaviors limitative of the intended change) Fig. 1.--Classification of obstacles to change 18 trative reform efforts in Brazil. Any critique should be useful in devising alternatives to the courses of action deemed inadequate. Thus, the closing chapter of this dissertation relates this critique to future reform attempts, in an effort to provide some basis for the elaboration of a theoretical model of administrative change strategy in Brazil. It is hoped that such extra polations into the future may contribute to the formulation of administrative change strategies in Brazil and, pos sibly, in the field of development administration. Major Analytical Concepts: Towards Models of Administrative Change Strategy This research is built around several constructs, among the most important of which are two models of change strategy. One reflects the strategic style which appears to have guided the attempts of administrative reform in Brazil— the instrumental model. The other will be referred to as the noninstrumental model. Both were conceived through the process of induction. However, such processes revolved around distinct referents, as the analysis of the formulation of the models reveals. By suggesting the basic 19 dimensions of any model of change strategy, the following section is an indispensable introduction to that analysis. Dimensions of Any Model of Change Strategy: Value. Content, and Action Any model of change strategy can be analytically reduced into three distinct components: (1) a valuational component, (2) a substantive component, and (3) an operational component. The valuational component refers to assumptions, beliefs, preferences, and attitudes that underlie a strategy. Whether the change agent perceives his role as value-neutral, rather than value-commited; whether he assumes that the prevailing behavioral pattern in the client system is or should be of a legal-rational, rather than of a particularistic, type;^ or whether he perceives cognitive, rather than valuational, obstacles, as the most deterrent ones, result in different strategies of change. Such values of the strategist, which are often implicit in ^This would imply, for instance, that one holding a position of command would take advantage of that posi tion for the benefit of the bureaucracy, and never for one's own personal interests. the change process, are ultimately reflected in the strategy itself. They are hereafter subsumed in the expression "strategy ideology." The substantive component refers to the actual content of a strategy. Data on this component are found mainly in the proposals or plans of change. Whether the style of a proposal is prescriptive, rather than suggestive whether, in the analysis of obstacles to the intended change, cognitive hindrances are considered more carefully and extensively than valuational ones; or whether the measures contained in the change proposal have been drawn from the experience of the client system itself (endogen- etic content), rather than from the experience of other systems (exogenetic content) ,-*-0 lead to important differ ences in the substantive content of a plan of change. This component is hereafter called "strategy content." l^The application of the notions of endogenesis and exogenesis to the substantive content of a change strategy was inspired in a discussion of endogenous and exogenous change by Riggs. Fred W. Riggs, Administration in Developing Countries— The Theory of Prismatic Society (Boston: Houghton Mifflin Co., 1964), pp. 38-42. 21 Finally, the operational component refers to characteristics that emerge only when a strategy is being implemented. Whether communication develops predominantly in a one-way, rather than two-way, process; whether the action of change centers on formal, rather than informal, relationships; or whether recruitment to the change group is based on a regimentative, rather than a voluntaristic, criterion are different modes of precipitating change. Those manifest traits of a model "in operation" are here after referred to as "strategy behavior," i.e., the actions developed by the change agent in order to accomplish his purposes. Observe that the three components of any model of change strategy— action, content, and value— are not dis tinct from each other in reality. Their separation is defensible only in abstract, for analytical purposes. Actually, the significance of action, content, and ideology for the characterization of a model of change strategy is likely to be unevenly distributed. By this is meant that one of the components, the valuational component, condi tions the other two. More precisely, the value system of the change agent bears decisively on his perception of 22 what is to be done (strategy content) and how it is to be done (strategy behavior). The reverse order of influence, from action to value, is not tenable.H The prominence of the valuational dimension of a strategy of change is the reason why, although the opera tional and the substantive components are not ignored, this study is basically concerned with the ideology of the reforms. Indeed, the analysis is almost totally directed to the implications of the ideology of the prevailing pattern of change strategy in Brazil, in view of the behavioral characteristics of the bureaucracy. The Ins trjBoental Mode 1 of Change Strategy The instrumental model was derived from two main referents— one, pragmatic, the Brazilian experience of administrative reform, and the other, theoretical, the Weberian conception of bureaucracy. The former was the 3 - Also, those components impose different diffi culties upon the researcher. For instance, many character istics of strategy behavior can only be detected by the participant observation technique, whereas strategy con tent can always be studied through content analysis; strategy ideology, on the other hand, might be identified through both a survey of the reform literature and inter views with the change agents. major source of the model. Indeed, on the basis of a cursory examination of the history of Brasil's civil ser vice, this author developed some tentative diagnoses of the reform movements. For one thing, there has been a vide gap between the proposed goals and the actual achieve ments of administrative reforms in Brazil.^ For another, the strategies whereby reforms have been attempted do not seem to differ significantly from each other; in some sense, there have not been strategies, but a specific pre vailing strategic pattern.^ This is suggested by some impressive regularities which not only are common to reforms in distinct areas such as materiel and budget, but also to successive reform efforts in one same area such as position classification.^ Should those diagnoses be J-^This point is fully explored in Chapter III. l^The demonstration of this point is one of the major concerns of Chapter 111. l^See, this regard: Kieber Nascimento, Classificacao de Cargos no Brasil. Vol. 13 of Serlede Administracao Comparada (Rio de Janeiro: Instituto Brasileiro de Ciencias Administrativas, 1962); also: Gilbert B. Siegel and Kieber Nascimento, "Formalism in Brazilian Administrative Reform: The Example of Position Classification," unpublished paper, June, 1965. (Mimeo graphed.) correct, the theoretical and pragmatic relevance of identifying and defining the characteristics of such a prevailing strategic pattern could not be ignored, for only after that identification and definition would it be possible to assess the adequacy or inadequacy of the strategy of the reforms, in view of the behavioral char acteristics of the Brazilian bureaucracy. As the reader might now anticipate, a major assumption underlay the author's whole reasoning, namely, that there might exist a considerable incongruency between the strategic patterns of the reforms and the behavioral patterns of the client system,if such a hypothesis is correct, one explanation for the failures of the reforms might well lie in this incongruency. It was under such an assumptional framework that this author turned to the Brazilian experience so as to abstract from it some characteristics which, combined together, might be taken as a model of the strategy that • ^Actually, that assumption became the basic hypothesis of this dissertation. Hence the need for identifying, on the one hand, the strategic pattern of the reforms, and, on the other, the behavioral profile of the Brazilian bureaucracy, so that they could be mutually con trasted afterwards. 25 has prevailed in the reform attempts. A first version of the model was thus developed, and after successive re visions, it came to the form presented in Fugure 2.^ The theory of bureaucracy had a complementary influence, particularly as a catalyst of concepts. For example, assuming that the change strategy of the instru mental model is hierarchy-oriented and also depends on behaviors oriented towards bureaucratic, rather than per sonal, ends, one can improve the communication of such an idea by saying that the strategy requires "functional position-orientedness."^-7 On the other hand, once one characteristic of the strategy was detected and abstracted into a concept, it would itself elicit insights about other possible characteristics. For instance, the focus of action of a given strategy being the formal, rather than the informal, structure is somehow suggestive that such a strategy depends on position-oriented behaviors. ^The demonstration that the reforms have followed the instrumental model of change strategy is the major purpose of Chapter III. 17For an extensive discussion of functionality and dysfunctionality of position-orientedness, see Chapter IV. 26 w C o •rt W c t Q J S • H a o ■ M ! 00 o r^i o < u T3 H b b ( U 4J r t v i V I c n 4 - 1 a < 1 ) VI e o u oo Q ) V I n ) V i VI M T7 0 •H ■d 0 1 PP >,. O T j I a i I jj I r t V i VI r n Characteristics of the Model 1. Authority patterns (which are required from the client system) 2. Time patterns (which are required from the client system) 3. Change scope 4. Language style 5. Obstacle mapping 6. Focus of action a. Functional position- orientedness in executive behavior b. Functional position- orientedness in expectations about executive behavior Optimizing-organizational temporal orientations Wholistic Prescriptive Cognitive impediments Formal elements Fig. 2.--The Instrumental model of change strategy 27 Therefore, the instrumental model emerged primarily from the Brazilian reality, while relying on bureaucratic theory for conceptualization. It must be observed that other theoretical referents also performed th^'function of catalyst of concepts in the formulation of the model. The notions of who 1 ism and gradualism, for example, were much inspired in Albert 0. Hirschman's view of development as a chain of disequilibria, rather than a balanced process of growth.The influence of such other referents will be duly explored in the next chapter.^ what is intended at this point is only to emphasize that, though bureaucracy was the most influential of the theoretical referents, it was not the only one; thus, the instrumental model should ■^Hirschman, The Strategy of Economic Development. Chapter IV. •^Still other inputs contributed to the formulation of the model. These, however, defy any scientifically reasonable attempt to evaluate their relative influence. Among such inputs are the author's interest in comparative public administration, especially regarding the issue of the transferability of techniques; several discussions with the members of his Dissertation Committee, particu larly with his Chairman, Dr. Frank P. Sherwood, who first suggested the theme of this dissertation; and even circum stantial factors like the availability of unused data on authority relationships in Brazil. not be called, or equated to, a bureaucratic model of change strategy. It is also critical for this matter to keep in mind that the major referent from which the model was inducted was a pragmatic one— the Brazilian experi ence of administrative reform— with the theoretical refer ents being used to sharpen the insights suggested by this experience. A Noninstruroental Model of Change Strategy The image of a counterpart for what has been the prevailing strategic pattern of the Brazilian reforms might be an insight-eliciting device of great usefulness in the conception of the instrumental model. Logically, the identification of A (the instrumental model) might be facilitated by reference to a non-A (a noninstrumental model). It was out of such a need that the noninstrumental model first emerged. Reworked several times, the model came to the version presented in Figure 3. The noninstrumental model shares with the instru mental one the inductive mode of elaboration. However, their formulatory processes differ as to the referents Analytical Dimensions 29. Characteristics of the Model 00 o O < u t) M s * . 00 ( U 4 J « j 14 4 4 CO 1. Authority patterns (which are required from the client system) Time patterns (which are required from the client system) No prerequisites No prerequisites 4J a o 4 J E S O o >. b O 0 ) 4-1 « s >4 4 4 co 3. Change scope 4. Language style 5. Obstacle mapping Gradualistic Suggestive Valuational impedi* ments M o f4 B A 0 ) « §3 4 1 4 - 1 c a 4 4 4J CO 6, Focub of action Informal elements Fig. 3.--A noninstrumental model of change strategy from which the models were inducted. While the instru mental model was derived primarily from an empirical referent— the Brazilian experience of administrative reform— the noninstrumental model emerged mainly from theoretical developments in the social sciences. The idea of the informal structure as the focus of action of a strategy, for instance, is a product of the human rela tions movement. The importance of a participative approach, whereby desired values come to be internalized by a group, led to the idea of contrapositing strategic models whose obstacle-mapping processes revolved around different sorts of obstacles— the instrumental model cen tering upon cognitive impediments, while its counterpart is directed more towards valuational and systemic barriers. The notion of a suggestive, rather than a prescriptive, language style, is also behaviorally-rooted; and the idea of gradualism is inspired primarily in Hirschman^ and Bogardus. 20see footnote 4. ^Emory S. Bogardus, "The Concept of Gradualism," Sociology and Social Research. XLIV, No. 2 (November- December, 1959), 119-24. 31 The behavioral-theoretical sources of the non instrumental model are more evident in regard to the strategy ideology of the model. This part grew out of the author's readings on the issue of preconditions for economic growth, as against the idea that a system can 11. . . begin to move forward as it is. because of what it is. and in spite of what it is."22 More precisely, after having conceived of the strategy ideology of the instru mental model, the author came to think of that model as an essentially prerequisite-based one. That is to say, a change attempted by means of the instrumental model of strategy requires given patterns in the dynamics of the client system, as preconditions for that change to suc ceed. In the case of authority relationships, for instance, the prevalence of functional position-oriented- ness is required. A comparable concept in the literature of economic development is that economic take off can occur only after certain preconditions exist,By ^Hirschman, Journeys Toward Progress, p. 6. Italics in the original. 23W. W. Rostow, The Stages of Economic Growth— A Noncomnunist Manifest (Cambridge: Cambridge University Press, 1960), especially Chapters III and IV. 32 contrast, there is the notion of a strategy of economic development not based on a primum mobile or on any sort of prerequisites. This notion was applied to the formu lation of the noninstrumental model of change strategy. Instead of being conditioned by certain factors, such a model would reflect a fundamental concern for the opera tional viability of the strategy under whatever set of characteristics might prevail in the client system. Thus, the noninstrumental model does not require any given pattern of authority relationships, or of temporal orien tations. Because of such "uncommitment," this model is much less precise than would be desirable. Also, the need which motivated the creation of the model, i.e., the need for an insight-eliciting device, makes for some con siderable differences in the relative importance, roles, and limitations of the two models, as it is argued in the coming section. ^Hirschman, The Strategy of Economic Development, especially Chapter I. 33 l i r e . Jteta &0leg and Limitations In social science, model building meane polemic building. Lacking the concreteness of models in the physical sciences, in which an atom can be represented as materially as a chair or a table, models such as those depicted in Figures 2 and 3 are bound to evoke unlimited dissensus. It is suggested that a comparison of the two models, from the viewpoint of their relative importance, roles, and limitations, may enlarge the "zone of accept ance" of those constructs. The points of comparison are complementary to each other. They seem to embrace most of the questions which are more likely to be elicited by the models. First, the two models are not antinomic to each other. That is, they are not mutually exclusive, much less extremes of a continuum. They are meant to depict, in abstract terms, two different— but not necessarily antinomic— modes of precipitating change. The noninstru- 25ihe majority of such questions was raised by the author's Dissertation Committee, in a discussion of the first draft of this chapter. mental model is not an anti-Instrumental model. It is sub mitted, however, that the instrumental model is more akin to a mechanical perspective of administration and, con sequently, to a lesser concern with the values that may prevail in the client system; noninstrumental models are more representative of a sociopsychological view of the administrative process, and, therefore, of a greater con cern with values as obstacles to change. In a gross com- t*ie tusLruuieiitttltst. change agent is an engineer, the noninstrumentalist an anthropologist. Second, the instrumental model is much more import ant for this research than its counterpart. This is so because of the different roles of the two models. While the instrumental model is intended to represent the actual strategic pattern which has prevailed in the Brazilian administrative reforms, the noninstrumental model here has only the accessory function of clarifying the concep tualization of the instrumental one. Thus, the central purpose of Chapter III is to demonstrate that the reforms in Brazil have followed the instrumental model. This model then becomes indispensable to the study, which is not the case of its counterpart. The difference of 35 importance and role of the two constructs is more clear when one considers that what the study purports to show is that the instrumental model has been followed, and not that the noninstrumental model has not been followed. Third, as a model, the noninstrumental has many more limitations than the instrumental, in that the former is much less specific than the latter. Indeed, there is only one instrumental model— the one abstracted from the 26 Brazilian experience*— but there can be, theoretically and pragmatically, thousands of noninstrumental models. It was in order to point out such a difference that the subheadings of the sections on the two models refer (1) to "the'1 instrumental model, and (2) to "a" noninstrumental model of change strategy. As a consequence of this, there are different requirements as to the internal consistency ^Obviously, it is possible, probable Indeed, that administrative reforms other than those considered in this study have relied on models which more or less resemble the instrumental model. If so, the applicability of this study will be widened. However, there is no concern with such other reforms here; no attempt is made to develop a most abstract form of the instrumental model which would embrace all possible variations of an instrument ally- oriented strategy. Rather, the point is to develop, if one may say «u>, the Brazilian instrumental model of change strategy. 36 of the two models. For instance, the instrumental model must have every one of its characteristics confirmed by an analysis of the Brazilian experience. That is, if only five of the six characteristics of the model were con firmed, the sixth one would have to be eliminated, so that all the dimensions of the model could be identified in the Brazilian experience. No attempt is made to probe into the internal consistency of the noninstrumental model, for the very reason that it actually covers a large variety of models within its context. For instance, a change strategy with all the characteristics of the noninstru mental model, except in regard to the focus of action, which would be the formal, rather than the informal, structure, might still be considered noninstrumentally- oriented. This would be even more the case if the formal and informal structures tended to converge upon each other, instead of being characterized by a formalistic relation ship. In sum, the two models are different from, but not antithetical to, each other; one is less concerned with values as obstacles to change than the other; one is indis*- pensable, the other is accessory; one is "finished,1 1 its 37 parts necessarily holding together; the other is highly diffuse, imprecise. These differences between the models imply also differences in the relevance of possible criti cisms to them. Criticisms to weaknesses of the noninstru- mental model should evoke much less concern (because of the relatively minor consequences of such weaknesses for the study as a whole) than those levied against the instrumental model. The Research Design^- Major Hypotheses This research centers on the two following major hypotheses; 1. Administrative reforms in Brazil have been patterned after an instrumental, rather than a noninstru mental, model of change strategy. 2. The behavioral patterns required by the strategy ideology of the instrumental model are incongruent with those which characterize the behavioral profile of 2?This design intendedly excludes the description of sampling, data processing, and statistical methodology. Such topics were deemed to be more adequately developed at the chapters for which they are relevant. 38 the Brazilian bureaucracy. Minor Hypotheses The characterization of the behavioral profile of the Brazilian bureaucracy is pursued through the develop ment and testing of several minor hypotheses. They focus on two of the most significant traits of the behavioral profile of any organization— patterns of authority and time. Thus, the following minor hypotheses will also be explored: In regard to authority patterns. 1. There is no significant difference in position*- orientedness of executive behavior between public strators in Brazil and the United States. 2. There is no significant difference in position-^ orientedness of expectations about executive behavior between public administrators* in Brazil and the United States. In regard to time patterns. 1. There is no significant difference between public administrators in Brazil and the United States in 39 regard to their perception of time as a scarce commodity. 2. There is no significant d?'.fference between public administrators in Brazil and the United States in regard to their concern for scheduling. 3. There is no significant difference between public administrators in Brazil and the United States in regard to their use of work hours for organizational, as against personal, purposes. Operational Definitions For the purposes of this study; Change agent. Change agent refers to a person or group who is deliberately attempting to effect change; the terms "change agent," "strategist," "reformer," and "change group" are interchangeable. The change agent may or may not be a member of the client system. Client system. Client system refers to a group, organization, or set of organizations which is the object of change. Instrumental model. Instrumental model refers to a change strategy (1) whose focus of action is on formal 40 elements; (2) which tends to a who1istic change scope; (3) is presented in a prescriptive language style; (4) antici pates cognitive impediments predominantly; and (5) requires the predominance of functional position-orientedness in the authority patterns and of optimizing-organizational orientations in the time patterns of the client system (see Figure 2). Noninstrumental model. Noninstrumental model refers to a change strategy (1) whose focus of action is on informal elements; (2) which tends to a gradualistic change scope; (3) is presented in a suggestive language style; (4) anticipates valuational impediments and incom patibilities predominantly; and (5) is not conditioned to any particular kind of authority and time patterns which may prevail in the client system (see Figure 3), Focus of action. Focus of action refers to the characteristics of the client system which are predomi nantly acted upon, in the attempt to change. On an ex pos t facto basis, such characteristics can be identified by analyzing documents of reform, newspaper interviews, and any literature about the intended change. 41 Formal elements. Formal elements refer to all legally prescribed or sanctioned relationships, both between organizational units and occupants of positions, in the client system. Sources of data on the formal elements are statutes, ordinances, edicts, decrees, regu lations, manuals of procedures, flow charts, and related documents. Informal elements. Informal elements refer to the actual patterns and nets of interaction among the person nel of the client system. The most common approach to measuring those patterns is by the use of sociometric techniques. Change scone. Change scope refers to the degree of inclusiveness of the change attempted, both vertically and horizontally. In the client system considered in the study, the horizontal inclusiveness of a change attempt is indicated by the number of ministries to be explicitly affected by that change; in the same way, vertical inclu siveness is indicated by the number of hierarchical levels (of one or several ministries) to be explicitly affected by the intended change. 42 t-Tholistie. Wholistic refers to a change scope which tends to encompass most of the horizontal and/or vertical dimensions of the client system. A change scope is said to be wholistic when it explicitly involves more than half of the ministries or more than half of the hierarchical levels of a ministry, or both. Gradualistic. Gradualistic refers to a change scope which tends to exclude most of the horizontal and vertical dimensions of the client system. A change scope is said to be gradualistic when it explicitly involves less than half of the ministries or less than half of the hierarchical levels of a ministry, or both. Language style. Language style refers to the degree of imperativeness conveyed in the wording of a change. The language style can be identified by analyzing publications by the change group, included, of course, the change proposals. Prescriptive. Prescriptive refers to a language style which is predominantly mandatory; solutions are "musts" to be pursued, little or no discretion being left 43 for alternative courses of action, should unanticipated conditions evolve. Suggestive. Suggestive refers to a language style which is predominantly recommendatory; solutions are '’ mights" to be pursued, much or conplete discretion being left for alternative courses of action, should unantici pated conditions evolve. Obstacle mapping. Obstacle mapping refers to the configuration of difficulties that are to be overcome in the change process. Such difficulties are often explicitly anticipated in the change plan, and their identification can be done by analyzing the literature about the intended change. Cognitive impediments. Cognitive impediments refer to obstacles that derive from a shortage of information or inadequacies of work methods and processes in the client system. Valuational impediments. Valuational impediments refer to obstacles that derive from assumptions, beliefs, attitudes, and values that underlie the behavior of individuals in the client system. 44 Authority patterns. Authority patterns refer to perceptions and expectations of individuals in the client system as to the exercise of authority by those invested in a given position. The authority patterns required by the strategy ideology of the administrative reforms in Brazil will be identified through analysis of literature as well as official documents on the reforms. The actual authority patterns of the client system will be identified through the use of questionnaires applied to a sample of that system. Position-orientedness. Fosition-orientedness refers to authority patterns which place more emphasis on position as against other factors such as competence, skill, and human relations, in commonly occurring adminis trative situations.28 ^^This definition was used in an empirical study by Mohammed Humayun Khan, "Authority Behavior: An Empiri cal Approach to Comparative Public Administration" (unpub lished doctoral dissertation, School of Public Administra tion, University of Southern California, 1965), p. 194. 45 Functional. Functional refers to authority pat terns which are oriented towards bureaucratic efficiency and organizational goals and interests, irrespective of personal influences. Dysfunctional. Dysfunctional refers to authority patterns which are oriented towards personal aims and particularistic interests, irrespective of their conse quences for the organization. Functional position-orientedness. Functional position-orientedness refers to the combination of the characteristics described in the definitions "Position- orientedness" and "Functional." Dysfunctional position-orientedness. Dysfunc tional position-orientedness refers to the combination of the characteristics described in the definitions "Position-orientedness" and "Dysfunctional." Time patterns. Time patterns refer to behaviors, perceptions, and expectations of individuals in regard to the temporal dimension of life, and especially to the allocation of their working hours to the fulfillment of 46 the duties and responsibilities that pertain to their position, in the client system. The time patterns required by the strategy ideology of the administrative reforms in Brazil will be identified through analysis of literature as well as official documents on the reforms. The actual time patterns of the client system will be identified through the use of questionnaires applied tp a sample of that system. (fctimizing-organizational temporal orientations. Optimizing-organizational temporal orientations refer to time patterns which are characterized by a high concern for scheduling and expeditiousness, and by the prevalence of organizational, as against personal, goals in the allocation of working hours, in the client: system. ■££rategV-ldeqlogjjy of the instrumental model. Strategy ideology of the instrumental model refers to a composite of (1) a prevailing functional position- orientedness in the perceptions of, and expectations about, the behavior of executives, and (2) a prevailing optimizing-organizational pattern of temporal orientation*» 47 Behavioral profile. Behavioral profile refers to perceptions, expectations and actual orientations of individuals related to their activities in the client system. The characterization of the behavioral profile of the Brazilian bureaucracy will center on two points: (1) the prevailing authority patterns of the client system; and (2) the prevailing time patterns of the client system. Contents of the Dissertation The next chapter is devoted to a brief survey of the literature, with the object of linking the two models to other developments in the theory and practice of planned change. That chapter completes the theoretical background of the dissertation. In Chapter III, the Brazilian experience of admin istrative reform is reviewed, and, on the basis of the literature on that matter, an attempt is made to demon strate (1) that a given strategic pattern has prevailed in that experience, and (2) that such a pattern has the characteristics of the instrumental model of change strategy. 43 Chapter IV is devoted to a discussion of the implications of authority patterns to the inducement of change. There, the theoretical foundations for an empiri cal study of authority relationships in Brazil are laid down. Two important traits of the behavioral profile of the Brazilian bureaucracy— authority relationships and temporal orientations— are the subject matter of Chapters V and VI, respectively. An attempt is made in those chapters to demonstrate that the behavioral characteristic required by the instrumental model of change strategy are antagonistic to those which characterize the behavioral profile of the Brazilian bureaucracy. Finally, Chapter VII synthesizes the rationale of the dissertation and suggests that one possible explana tion for the failures of the Brazilian reforms lies in the conflict between the requirements of the strategy ideology of the instrumental model and the behavioral realities of the client system. Based upon the lessons of the past, some suggestions are made in regard to the strategy of future reform efforts. CHAPTER II TOWARDS MODELS OF ADMINISTRATIVE CHANCE STRATEGY: CONTRIBUTIONS OF THE LITERATURE ON PLANNED CHANGE The main purpose of this chapter is to reinforce the theoretical foundations of the instrumental and non instrumental models, by relating them to other endeavors in planned change. It is hoped that, as a result, some of the concepts of those models are clarified and better understood, Goncont&itantly, an effort will be done to dis close the limitations of the literature in regard to model devising and operationalization of models of change strategies (as distinguished from models of change). The methodology to be used is a survey of the per tinent literature, with a focus on pieces which may be relevant for the clarification of the Concepts which com prise the two models. The most elaborate application of 49 50 this methodological approach is by Garth N. Jones,1 assisted by the members of his Comparative Administration Seminar, at the University of Southern California. On the basis of a tenfold classification of change strategies,2 Jones and his students inventoried and classified 190 cases on planned change, using the technique of content analysis. They were able to buttress with empirical data every one of the conceptual categories of their proposed strategies. In order to facilitate the sharpening of its focus while not ignoring completely some important contributions, this survey is developed in two stages. First, the litera ture is briefly analyzed from the perspective of the level of systemic complexity at which one deals with change, i.e., whether the client system is an individual or an 1-Garth N. Jones (ed.), Planned Organizational Change— -A Set of Working Documents (Los Angeles: Center for Research in Public Organization, School of Public Administration, University of Southern California, 1964). While Jones focused more on empirical works and encompassed several types of change (community change, economic devel opment, et cetera), this survey is directed more to theoretical contributions in organizational change. ^That classification was originally developed by Aslam Niaz, "Strategies of Planned Organizational Change" (unpublished doctoral dissertation, University of Southern California, 1963). 51 international system. This perspective allows for the identification of the main dimensions of the literature, making it possible for the researcher to exclude (as well as justify such an exclusion) those contributions whose usefulness for the purposes of this survey is rather limited. The first stage of the survey is, therefore, a Justified exclusion, rather than inclusion, of materials. In the second stage, the focus is sharpened on contribu tions which are more directly related to the instrumental and noninstrumental models of change strategy, which appear together in Figure 4. Levels of Change: From Individual to Social change may be dealt with at several levels of systemic complexity, the unit of analysis varying from the individual to international groups. Attempts at theory building in planned change often fall to consider the extent to which generalizations about a given level are also applicable to other levels of systemic complexity. The best illustration of an effort to develop generaliza tions for several levels of change Is provided by Lippitt Characteristics Instrumental Model Noninstrumental Model /tn 1 C 0 ■H <0 a 1 iH a Strategy Ideology 1. Authority patterns (which are required from the client system) 2. Time patterns (which are required from the client system) a. Functional position-orientedness in executive behavior b. Functional position-orientedness in expectations about executive behavior Optimizing-organizational temporal orientations No prerequisites No prerequisites r-1 <4 o •H U >» r~4 ta a < Strategy Content 3. Change scope 4. Language style 5. Obstacle mapping Wholistic Prescriptive Cognitive impediments Gradualistic Suggestive Valuational impedi ments Strategy Behavior 6. Focus of action Formal elements Informal elements Fig. 4--The two models of change strategy 53 and his associates.3 They focused their work on four "levels of problem-solving effort,"4 namely, the personal ity, the group, the organization, and the community, and theorized about the change agent's diagnostic orientations and role, the phases of planned change, and related points, with a view on applying their inferences to the four levels of change. The book by Lippitt and hie associates represents a most valuable contribution to the field. First, it served a catalyst function, by bringing together the experiences of professionals in many different areas of the practice of planned change— psychiatrists, social workers, marriage counselors, mental hygienists, confer ence experts, leadership trainers, productivity analysts, labor-management specialists, community self-survey con sultants, community-council organizers, and others. Second, it represented a benchmark in the terminological development of the discipline. Concepts such as client system, change agent-, change forces, phases of change, and ^Ronald Lippitt, Jeanne Watson, and Bruce Westley, The Dynamics of Planned Chance (New York: Harcourt, Brace & World, Inc., 1958). 4Ibid.. p. 5. 54 methods of change gained a certain consistency of meaning to professionals who are to operate with, or theorize about, them; such concepts have since then facilitated the communication of ideas in planned change. Third, though not explicitly using the expression "model of strategy," the authors provided an excellent treatment of methods of change which are directly related to the models of change strategy used in this study.5 As far as limitations are concerned, the book also deserves certain remarks. First, it left out of consideration changes at the national and international levels. Second, it also excluded "... a vast swath of change phenomena such as technological change, changing health habits, changing technical skills., changing modes of economic behavior, and the like."6 Third, it only considered change attempts (1) in which the change agent is an outsider, i.e., does not belong to the client system; and (2) in which the relationship between the external change agent and the client system is a ^This part of the book is examined under the sec tion "Planned Organizational Change: Towards Model Building," below. ^Lippitt, Watson, and Westley, op. cit„. p. 15. 55 voluntary one.^ Finally, the authors' effort at general izing for the four levels of change was only partially successful. As they acknowledge in their discussion of diagnostic orientations: V7e are tentatively proceeding upon the assumption that each diagnostic orientation is equally applicable to units of all sizes, including the individual, the small group, the institution or organization, and the community. In selecting illustrations for the chapters which follow, we have tried to find examples from cases involving as many of the four kinds of client systems as possible. Unfortunately, ouy success has been only partial. Some of the orientations seem to apply more directly to work with one kind of system than to work with the others. Nevertheless, we feel that an attempt to imagine what a given approach will mean when applied to each of the four kinds of client systems, even though the actual case materials are lacking, may be both stimulating and rewarding. We invite our readers to lend us their imaginative cooper ation at the points where our restricted and somewhat arbitrary list of examples breaks off.8 The idea of structuring a discussion of planned change on the criterion of levels of systemic complexity ^Ibid., Observe that this is the most important limitation of the book, for this dissertation, since in the Brazilian experience the change agent has usually been a member of the client system; moreover, during the dicta torship, the relation between change agent and client system did not always rest on a voluntary basis. ^Ibid.. p. 22. Italics added 55 9 was also attempted by Amitai and Eva Etzioni. Though much less exhaustive than Lippitt and associates', the Etzionis1 work is more inclusive, starting with the small est unit of all social collectivities, the role, and suc cessively moving to the face-to-face group, the organiza tion, racial groups, the society, and the international system. Planned change at the personality level is the domain of psychiatrists and psychologists. Of all levels, this is the one with the most voluminous literature. It is, however, only indirectly related to the main concern of this survey, among other reasons because, by their naturally restricted area of intervention, studies of individual change are not likely to involve the gamut of variables that affect change in large organizations. The psychiatrist usually confines his analysis of environmental variables— those which are external to the individual— tc the family and, sometimes, to the work relationships of hi3 client. There is a great difference between the degree of 9Amitai and Eva Etzioni (eds.), Social Change— Sources. Patterns, and Consequences (New York: Basic Books, Inc., Publishers, 1964), Part V. 57 complexity of the individual’s and the large organiza tion’s environments. To say the least, the task of devis ing conceptual frameworks whereby the forces operating in those two environments would be related to each other is yet to be done. On the other hand, those contributions on individual change which try to account for more complex constellations of variables tend to lose the conceptual operativeness that might make them useful to model build ing. Usually, such contributions are of a culture-and- personality kind; here, psychologists share with anthro pologists the concern for change. Examples of contribu tions of this sort are those by Erikson,10 Newcomb,11 10Erik Erikson, Childhood and Society (New York: Norton, 1950). The concept of identity crisis, central to Erikson's contribution, has been brilliantly extrapolated to the national level of change by Lucien W. Pye, in his Politics. Personalitv, .and. Nat ion-Building jl Burma’s Search for Identity (New Haven: Yale University Press, 1962). ^Theodore M. Newcomb, Personality and Social Change (New York: Dryden Press, 1943). 58 Piaget,12 sherif,13 a^d Lewin,!^ all psychologists; and by Linton,15 Mead,16 and Kardiner,!? anthropologists. The next two levels of systemic complexity— the face-to-face group and the organization— are of direct relevance for the purposes of this survey. Since a discus sion of large organizational change will necessarily involve studies of groups, these two levels of change will not be distinguished from each other, in this survey. Rather, they will be explored as a common category, in the section, "Planned Organizational Change: Towards Model Building," below. 12Jean Piaget, The Moral Judgement of the Child (New York: Harcourt, Brace & Co., 1932). n JMuzafer Sherif, The Psychology of Social Norms (New York: Harper & Bros., 1936). l^Rurt Lewin, A Dynamic Theory of Personality (New York: McGraw-Hill Book Co., 1935). l^Ralph Linton, The Cultural Background of Person ality (New York: D. Appleton-Century Co., 1945), l^Margaret Mead, "The Implications of Culture Change for Personality Development," American Journal of Ortho-Psychiatry. XVII (1947), 633-46. I?Abram Kardiner, The Individual and His Society (New York: Columbia University Press, 1939). 59 When one turns to change at the societal, national or international levels, the difficulties of operationali zation resemble those of the culture-and-personality type of studies. Here again, the clusters of variables which affect change differ from those relating to organizational or group change. The external environments of an organiza tion and of a society, for instance, involve very different degrees of complexity; hence, the much greater difficulty of building models of strategy for the changing of a society, as compared with organizational change; hence, also, the tendency of "societal” and "cultural" sociolo gists toward grand theory,and of organizational sociologists toward operationalism. ^This expression was created by C. Wright Mills as an opposite to what he calls "abstract empiricism." In oversimplified terms, "grand theory" means theory not sub mitted to testing against reality; and abstract empiricism refers to data searching unguided by any theoretical frame work. From a scientific standpoint, both practices are condemned by Mills. See: C. Wright Mills, The Sociologi cal Imagination (New York: Oxford University Press, 1959), especially Chapters II and III. 60 Models of Change: Classical and Modem Theories Both c l a s s i c a l ^ anti modem theories of societal or cultural change were constructed under a given model. For Spencer,20 societies change by linear evolution, from simpler to more and more complex structures and functions, passing through three basic stages: (1) the theological and military epoch, (2) the metaphysical and juridical epoch, and (3) the scientific and industrial epoch. These are also the three historical stages of civilization. Comte21 developed a similar model of social change; for him him, human thought (and the social organization as well) moves from a theological to a metaphysical to a positive stage. Spencer and Comte's pattern of linear ascent is in contrast to the views of Spengler, Toynbee, Marx, and ^The following analysis of classical theories— from Spencer to Toennies— is based on Amitai and Eva Etzioni, op. cit.. Part I. Since the reader's judgment might be facilitated if the original sources were indi cated, such an indication is done for each one of the authors included in the analysis. ^Herbert Spencer, Sociology. Vol. I (New York: Appleton & Co., 1892). 2lAuguste Comte, System of Positive Polity. Vol. IV (London: Longmans, Green & Co., 1877). 61 Toeimies. Spengler's^ model of social change is circular, not linear. He sees cultures moving through a life cycle, much like that of man, from birth to death. Observe that underlying all such theories are the ideas of evolution and determinism, i.e., they account for neither signifi cant retrocesses nor the inability of a society to move toward a next and predetermined stage. T o y n b e e23 refined Spengler's circular model, by introducing the notion of shortcuts, as well as of indeterminism in the life cycle of cultures. That is, some cultures achieve maturity with out necessarily passing through the same stages through which other cultures had to pass. Moreover, some cultures are abortive, others die before maturity. Thus, Toynbee attributes to the social forces a more decisive role in the determination of a society's destiny. From a Marxist^ perspective, social change is also bound to historical 22()swald Spengler, The Decline of the West (New York: Alfred A. Knopf, Inc., 1926). ^Arnold J. Toynbee, A Study of History, abridged by D. C. Somervell (6 vols.; New York: Oxford University Press, 1946). 2^Karl Marx, A Contribution to the Critique of Political Economy (Chicago: Charles H. Kerr, 1904). 62 determinism, of which the classless society would be the last stage. For Marx, however, evolution is to proceed not through a linear pattern, but rather in a back-and- forth fashion generated by the class struggle. Clearly, Marx assumed a sharp dichotomic class stratification as a necessary consequence of industrialization, an assumption that has often been dismissed by reality. Finally, Toennies^ represents a return to the linear model— societies evolve from Gemeinschaft to Gesellschaft. the former oriented by sacral, status, and particularistic criteria, the latter by secular, contractual, and univer- salistic patterns. Toennies may be taken as the link between the mod e m and the classical theories of social change, and between both and the organismic view of society held by Spencer, in the nineteenth century. Indeed, modem con tributions like Parsons' pattern variables represent a crystallization of concepts already examined by Toennies, more than the introduction of new constructs in the ^ F e r d i n a n d Toennies, Community and Society— Gemeinschaft and Gesellschaft. trans. and ed. Charles P. Loomis (East Lansing, Mich.: The Michigan State Univer sity Press, 1957). 63 discourse of sociology. Affectivity pnd affective neutral ity, self-orientation and collective orientation, univer- salism and particularism, and diffuseness and specificity are, in fact, the differentiators between Toennies1 Gemeinschaft and Gesellschaft. As recognized by Parsons: As those of you familiar with some of my own writ ings since the Structure of Social Action know, for some years I have been "playing" with a scheme of what I have found it convenient to call "pattern variables" in the field of social structure, which were orieinallv. derived bv an analytical breakdown of Toennies1 Gemeinschaft-Gesel1schaft pair into what seemed to be more elementary components. This yielded such dis tinctions as that between universalism . . . and particularism . . . or to take another case, between the "functional specificity" of an economic exchange relationship and the "functional diffuseness" of marriage.2o These ideas are basic, in turn, to many other more recent theories dealing with social change in its broadest sense. This is the case of Redfield's folk s o c i e t y ,27 0f 96 * Talcott Parsons, Essays in Sociological Theory (rev. ed.; New York: The Free Press of Glencoe, 1954), pp. 359-60. Italics in the original. Parsons attributes the fifth pattern variable, ascription-achievement, to Ralph Linton (see p. 360). ^Robert Redfield, "The Folk Society," The American Journal of Sociology. LII (January, 1947), 293-308. 64 Sjoberg's feudal and folk societies,28 and 0f Riggs' fused, prismatic and diffracted societies.29 Theories on specific social subsystems were also influenced by Toennies' concepts. Such an influence can be seen, for instance, in Fye's model of a transitional politics,30 31 and in Almond and Coleman's models of political systems - both based on the degree of structural-functional specifi city achieved by a given system. Other models exist which revolve around different referents than the diffuseness-specifity continuum. This is the case of constructs such as those developed by La Piere, Apter, McClelland, and Hagen. After attempting to demolish traditional theories of social change based on a deterministic evolutionism or ^Gideon Sjoberg, "Folks and 'Feudal' Societies," The American Journal of Sociology. LVXII (November, 1952), 231-39. 2^Fred W. Riggs, Administration in Developing Countries: The Theory of Prismatic Society (Boston: Houghton Mifflin Co., 1964), especially Chapter 1. 30pye, op. cit.. Chapter II. ^Gabriel A. Almond and James S. Coleman (eds.), The Politics of the Developing Areas (Princeton: Prince ton University Press, 1960), especially the "Introduction" by Almond and the "Conclusion" by Coleman. 65 on particularistic causality (such as geographical, technological, or ideological determinism), La Piere3^ presents his three models of change and stability. The first, stable congruence, refers to: . . . A social complex or other unit within which the various parts or other elements are functionally congruent so that the system is not likely to be modified by forces from within that unit— i.e., by those who are involved in its operation— although there is always the possibility that its operation may be disturbed by external factors of a social, physical, or biological character.33 The second model, static congruence, is ". . .a condition in which incongruences exist within a social system or social unit and are perpetuated over a consider able period of time.'’34 Finally, the third model, dynamic incongruence, resembles the concept of open society, and is more facilitative of change. In La Piere's words: When the characteristics of the social system are such that the psychological tensions generated by incongruence between functionally interdependent social elements tend to be directed toward a modifica tion of those elements, rather than an exploitation 32Richard T. La Piere, Social Change (New York: McGraw-Hill Book Co., 1965). 33Ibid.. pp. 89-90. 34ibid.. p. 93 66 of them, a condition of what may be termed dynamic incongruence exists.35 It does not require much thought to ascertain that such models are not models of change at all. They are models of social structures, whose referents are the social circumstances of a system, at a given moment. At best, La Piere is describing characteristics of client systems, not how to intervene in them. And even so, his description defies any attempt of operationalization, a condition which is essential to model building, at least as it is dealt with here. Apter36 abstracted the political strategies which have been used in the new nations to induce technological change and economic development into three types. He calls the first type a mobilization system, examples of which are Mali, Ghana, and Guinea. These countries "... have chosen to mobilize their political energies and 1*1 HI I I — — — — — — 11 ■ I I W M I H l l „ .............. » ■ I .1— « MH— .m n — ■-! I — — ,..!■■■ ■ , l t ' ■ 35Ibid.. p. 99. ^^David E. Apter, "System, Process, and the Poli'» tics of Economic Development," in Industrialization and Society, ed. Bert F. Hoselitz and V7ilbert E. Moore (Mouton: UNESCO, 1963), pp. 135-58. These strategies are developed at greater length in Apter's recent book, The Politics of Modernization (Chicago: University of Chicago Press, 1965). 67 resources for a grand assault on poverty, ignorance, and backwardness."37 The second type, the reconciliation system, is characterized by an effort to use economic development as a factor of integration of the system, such as the case of Nigeria. The last type is the modernizing autocracy, in which "... change is filtered through the medium of traditional institutions— i.e., innovation it self is traditionalized and rendered compatible with traditional institutions."®® The first two of Apter's types are of no applica tion to the purposes of this study. The last type, how ever, may be of some value for the building of models of change strategy which, instead of considering the over coming of tradition as a requisite to change, would use traditional elements as leverages of change. In the case of the instrumental model, for instance, certain behavioral patterns in the client system are prerequisites for the strategy to succeed. Contrariwise, noninstrumental models do not rely on prerequisites, but try to use the prevailing 37Ibld.. p. 139., 38H>id. characieristics of a system in order to change it. Actually, this is the keenest difference between the two models, from a strategic standpoint. David M c C l e l l a n d ^ and Everett Hagen^O have come much closer than any of the authors discussed in this chapter to the issue of change strategy. For the first, the strategy should concentrate on a motivational factor, the need for achievement. Therefore, instead of transfer ring solutions or techniques, the change agent should try to develop a high n, achievement in the client system.^ Though of considerable value, McClelland's contribution is actually an application to large complex economic systems of what many sociologists and sociopsychologists have been repeating for a long time in regard to organizations, namely, that without motivation, change is not likely to occur. Thus, it seems more desirable to examine the ^David c. McClelland, The Achieving Society (Princeton, N.J.: D. Van Nostrand, 1961), especially Chapter X. ^Everett E, Hagen, On the Theory of Social Change How Economic Growth Begins (Homewood, 111,: Dorsey Press, 1962). ^McClelland deals with societies as a whole, with a focus on the economic subsystem. Also, his concern is with change in the developing nations. 69 literature on organizational and group change, which is done in the coming section. It might be argued that McClelland specified a particular kind of motivation, the need for achievement, nevertheless, his theory is still in need of much more empirical support. It may well be, as Hirschman suggests, that the difference between devel oped and developing nations is not in the abundance of high 31 achievers; The sudden outbreak of industrial entrepreneurship in Pakistan after the partition, in the Middle East during World War II, and again in Latin America since the thirties confirms the view that underdevelooed countries mav harbor a "reserve array1 1 of entrepreneurs that are as achievement motivated as anv Puritan ever S2 SS.* 2 The difference may lie in that high n achievers in developing nations may not have the same capacity to enlist cooperation for joint undertakings as their counterparts in the developed nations have had.43 If it is so, it may be strategically more efficient to try and infuse coopera tive ability, sense of teamwork, rather than need for achievement, in the developing nations. 42Albert 0. Hirschman, The Strategy of Economic Development (New Haven and London: Yale University Press, 1958), pp. 3-4. Italics added. 43ibid., pp. 16-20. ;o Everett Hagen also calls on a motivational factor for the unleashing of change. For him, the withdrawal of status respect is a main elicitor of the change process.44 The usefulness of his theory, however, is greatly impaired by its pragmatic difficulties. Since the reaction of those whose status is withdrawn is likely to bring about concrete results after two or three generations only, one would hardly use such a strategy to change a public bur- eaucracy. At best, this would imply the preparation of a strategy of change to be applied some thirty years afterwards. 5v the Wav of Conclusion As suggested by this cursory review of the litera ture from the perspective of the levels of systemic com plexity at which one deals with change, the contributions centering on levels other than those of groups and organi zations are of a rather limited value for the clarifica tion of the two models used in this study. Writings on individual change (except those of a culture-and-personal- ity kind) tend to present a degree of operationability 44uagen, o p. cit.. pp. 185-236 71 that makes them helpful to model building; also, when the change agent is a psychiatrist or a psychologist, one has the best example of planned, rather than accidental or fortuitous, change. However, analyses of individual change tend to exclude the variables and combinations of variables that usually affect change in large organiza tions. Thus, the usefulness of such analyses for the pur poses of this survey is greatly reduced. At the other extreme, studies about societal, cultural, or international change involve a much more com plex constellation of variables than those related to organizational change. This makes it very difficult, "economically1 1 unrewarding indeed, to seek and find in such studies support for building models of organizational change. In order to achieve some degree of generalizabil- ity, analyses of societal change tend to become grand theories, thereby loosing the operationality which is indispensable to model building. Also, many of such analyses do not deal with planned change at all. It seems indefensible to say that Spengler's life-cycle model or even Riggs' fused, prismatic, and diffracted models consti tute models of planned change, much less models of change 72 strategy. They describe stages of change, change which often has been more unintentional, even unconscious rather than deliberately planned. They may be considered models of change, not of change strategy. Because they are addressed primarily to describing change (which characteristics tend to be exhibited by a system when it reaches a given stage of change or development; which characteristics differentiate the several stages of devel opment from each other; et cetera), rather than to engen dering it (how to move a system from one stage to another), they result in descriptive, not in inferventive, models of change. Even when focusing on the inducement of change, contributions of this sort fall far from modeling strate gies of action and defining their components, as is the case with the instrumental and the noninstrumental models of change strategy. This section may well be closed with some words by Irwin T. Sanders. Though they are directed to grand theories, thereby not applying to some of the pieces examined above, such words express the general feeling of the reviewer who looks for guides of action or strategies of change: 73 1 find such theories intriguing and stimulating. They jolt me out of all complacency, but unfortunately they seldom give me any guide to action or any useful suggestions about the kinds of concrete studies of social change in which 1 become involved from time to time, whether it be the analysis of changes in social work as a profession or a community development pro gram in Indonesia.45 After such considerations, we may now turn to literature dealing with organizational change. Model Building This part of the literature survey focuses on change at the organizational level. Exceptionally, one or another contribution directed at other levels of systemic complexity than those will be examined. Uppltt, Watson and Vestlev's Contribu tion: Diagnostic Orientations The concept of diagnostic orientations is very valuable for the clarification of the instrumental and the ^->lrwin T. Sanders, "Approaches to Social Change," Education for Social Work. 1960 Proceedings, Eighth Annual Program Meeting, Council on Social Work Education, New York, New York, pp. 3-23. Address delivered in Oklahoma City, Oklahoma, January 20, 1960. Cited in Warren G. Betmis, Kenneth D. Beime, and Robert Chin, The Planning of Change (New York: Holt, Rinehart & Winston, 1961), pp. 401-402. 74 noninstrumental models of change strategy. That concept has been explored by several authors. It is basic to the rationale of Lippitt, Watson, and Westley's work,4* * and also to pieces written by Chin, Bennis, and Gouldner, as will be shown. Lippitt and his associates assert that: . . . Every change agent makes certain assump tions , either consciously or unconsciously. These pertain to (1) the nature of the client system, (2) the process by which the client system got into trouble, (3) the nature of the trouble, (4) the pro cesses which will lead to an amelioration of the trouble, and (5) the ways in which he himself can contribute to bring about the desired change. These are the assumptions which define hia approach to the task of helping with planned change The first three assumptions comprise what Lippitt calls the "diagnostic orientations" of different change agents.4® This concept is very similar to what has been defined here as the ideological component of any change strategy, namely, assumptions, beliefs, preferences, and attitudes that underlie a strategy; in one word, the ^Lippitt, Watson, and Westley, op. cit.. p. 22 4^Ibid.. Italics added. 48Ibid. 75 valueb and biases of the change agent, which are ultimately AQ reflected in the strategy itself. Let us assume, for instance, that two change agents have different diagnostic orientations about the process by which a given client system got into trouble, as well as about the nature of such trouble.50 if the diagnosis of change agent A about those points rested on factors such as shortage of tech nical skills, inadequate coordination, deficiencies of information, and lack of data; and if, on the other hand, the diagnosis of change agent B centered on points such as attitudes, habits, and values tdiich prevent a better functioning of the system, the change strategies to be devised by A and B would likely be quite different from each other. It is so because the ideological components of such strategies would differ from each other. Also, as suggested before, the ideology of a strategy has reflec tions on both the content and the actual operations of the change agent. For instance, A would more likely focus his action on the formal structure of the client system, ^See Chapter I, pp. 19-20. 50 See quotation from Lippitt, Vat son, and Westley, footnote 47, numbers 2 and 3 above. 76 whereas B would target informal relationships. Also, in regard to content, the strategy devised by A would more likely stress cognitive Impediments, for, in his obstacle- mapping process, A would be more concerned with shortage of data or inadequate lines of coordination, rather than with valuational impediments, as it would be in the case of B. Thus, A would more likely devise a change strategy similar to the one represented in the instrumental model, whereas B's strategy would conform more to a noninstru mental model.^ Actually, the book by Lippitt, Watson, and Westley rests upon the two general dimensions of planned change, diagnosis and prognosis. Though they do not put it in such terms, out of the five assumptions that they developed for any change agent, the first three are diagnostic orienta tions; the last two52— we would complete— are prognostic orientations. In broad terms, their diagnosis and prog nosis correspond to strategy ideology and strategy r behavior, in the instrumental and noninstrumental models See Figure 4. S^see footnote 47. 77 of change. It must not be forgotten, however, that it is often impossible to determine where diagnosis ends and prognosis starts; in this sense, it may be more meaningful to consider strategy ideology as encompassing all five assumptions developed by Lippitt and his associates, i.e., both diagnostic and prognostic orientations. Chin1s Three Models The concept of diagnostic orientations was used in a more elaborate way by Chin,^ with a view on developing models of change as explicit and specific as possible. Actually, this contribution is the one which comes closest to the idea— as well as to the form given to it— of models of change strategy which directs this dissertation. Chin's paradigm is reproduced in Table l.**4 The distinction between models of change and models of change strategy, which is essential to this study, and which has been the reason for eliminating from ^Robert chin, "The Utility of System Models and Developmental Models for Practitioners," in The Plannine of Change, ed. Warren Bennis, Kenneth D. Benne, and Robert Chin (New York: Holt, Rinehart & Winston, Inc., 1961), pp. 201-14. 54Ibid.. p. 213. TABLE 1 ASSUMPTIONS AND APPROACHES OF THREE ANALYTIC MODELS Assumptions and Models of Change AnDroaches to: Svstem Model Developmental Model Model for Chaneine Content Stability Structural integration Phases, stages Unfreezing parts Change Derived from structure Constant and unique Induced, controlled Causation Source of change Structural stress Nature of organisms Self and change agent Causal force Tension reduction Rational choice Goals Direction Emergent Ontological Deliberate selection Sec by "Vested interests" Collaborative process Intervention Confronting symptoms Stresses, strains and tensions Discrepancy between actuality and potentiality Perceived need Goal of intervening Adjustment, adaptation Removal of blockages Improvement Change Agent Place Outside the "target" system Outside Part of situation Role External diagnoser External diagnoser Participant in here and actor and actor and now ■**1 00 79 this survey a great bulk of the literature, is often unrecognized. Indeed, the researcher only found an expli cit recognition of that distinction in two works, one of them by Chin: The social scientist generally prefers not to change the system, but to study how it works and to predict what would happen if some new factor were introduced. So we find his attention focused on a "theory of change." of how the system achieves change. In contrast, the practitioner is concerned with diagnosis: how to achieve understanding in order to engage in change. The practitioner, therefore, has some additional interests; he wants to know how to change the system, he needs a "theory of changing" the system.^ In a strict sense, the system and the developmen tal models developed by Chin are still models of change, not of change strategy. Only his model for changing con stitutes a model of change strategy in the sense of the instrumental and the noninstrumental models, as it becomes clear from Chin's own words: It [the model for changing] incorporates some elements of analyses from system models, along with some ideas from the developmental model, in a frame work where direct attention is paid to the induced forces producing c h a n g e .-56 ^Ibid.. p. 211. Italics added. The other work, by Bennis, will be reviewed later. ~ * * *Ibid.. pp. 213-14. Italics added. 80 In general terms, however, it may be said that some of the assumptions and approaches of Chin* s system model are shared by the instrumental model, while the developmental and the "model for changing" models have certain resemblances to the noninstrumental model used in this study. Such a comparison must be taken in the broad est possible terms; also, it is not a point-by-point com parison, but one which is only acceptable in regard to certain parts of Chin's models. Still a further prefatory note: Chin does not operationalize the components of his models, as is done with the instrumental and the nonin strumental constructs. Therefore, it is often impossible to grasp the exact meaning of his categories, so that one has to assume, to guess indeed, what is meant by them. As he himself says, "The models for changing are as yet incompletely conceptualized."57 Also, very responsibly, Chin warns against the incompleteness and vagueness of his categories: We have varying degrees of confidence in our cate gories, but, as the quip says, we construct these in order to achieve the laudable state of "paradigm 57Jbii-. p. 21* 81 lost." It is the reader's responsibility to help achieve this goal158 With all such notes of caution in mind, a compari son between Chin’s constructs, and the instrumental and noninstrumental ones may now be undertaken. For the sake of brevity and clarity, the comparison is first done synoptically, by the reproduction of the comparable parus of the models in a tabular form. The reader will cer tainly need to refer back to Table 1 and Figure 4 in order to understand the comparison which is displayed in Table 2 . As indicated in Table 2, the developmental model and the model for changing tend to deal with parts, whereas the system model visualizes the whole; hence, there is a possible resemblance of the first two with the noninstrumental model, whose change scope is gradualistic; and of the latter with the instrumental model, whose . change scope is wholistic. In regard to focus of action, it is not clear, in Chin’s system model, whether the structure to which he refers is the formal or the informal one. However, one 53ibid TABLE 2 A COMPARISON BETWEEN CHIN'S MODELS AND THE INSTRUMENTAL AND NONINSTRUMENTAL MODELS OF CHANGE STRATEGY Models System Instrumental Noninstrumental Developmental Model for Changing Approaches sta bility through structural Integration Change scope: Whollstic Change scope: Gradualistic Approaches sta bility through phases, stages Approaches stabil ity through unfreezing of parts Assumes change as derived from structure Focus of action: Formal elements Focus of action: Informal elements Source of change is structural stress Language style: Prescriptive .............................. ... ....................... Language style: Suggestive . .................... Goals of change are set by collaborative process 00 to 83 may assume that Chin is talking about the formal structure when he asserts; A system model assumes that organization, inter dependency, and integration exist among its parts and that change is a derived consequence of how well the parts of the system fit together. , . .59 Finally, in respect to language style, the model for changing may be compared to the noninstrumental model insofar as, in the former, goals are set by a collabora tive, process, which rules out any prescriptive language style and is more in conformity with the suggestive language style of the noninstrumental model. In corroboration of this point, one might quote Chin again: "The direction [in the model for changing] to be taken is not fixed or 'determined' but remains in large measure a matter of 'choice' for the client system. As suggested before, this comparison is to be con sidered in the broadest possible grounds. It is no more than an effort to identify points of contact-— not of com plete correspondence at all— between the models used in 59it>ld.. p. 212. Italics added. 60Ifeii-. P- 2U. 84 this study and a given piece of the literature. It is hoped that this kind of effort may both clarify the con tent of the instrumental and the noninstrumental models, and make evident the scarcity— almost nonexistence— of contributions which are directly concerned with the modeling of change strategies, or ’ ’ models for changing." Bennis:__gjgfrE Types. Change Processes In a two-page article, Warren G. Bennis61 develops an eightfold typology of change processes. His purpose was to clarify the distinction between planned and other forms of human change. His paradigm is not, therefore, related to the present study in a direct way. However, one of Bennis' eight types of change (planned, indoctri- national, interactional, socialization, technocratic, coercive, "natural," and emulative), the technocratic change, has certain characteristics that suggest an instrumentally-oriented strategy, to wit: 61Warren G. Bennis, "A Typology of Change Pro cesses," in op. cit.. ed. Warren Bennis, Kenneth D. Benne, and Robert Chin, pp. 154-56. 85 The use of technocratic means to bring about change relies solely on collecting and interpreting data. Technocratic change follows primarily an "engineering model": the client defines his difficul ties as deriving from inadequate knowledge and assumes that his lack of knowledge is accidental or a matter of neelect— not something that is functionally part of him. The technocrat colludes in this assumption and merely makes and reports his "findings,"62 The excerpt above deals almost exactly with the concept of obstacle mapping, which is applied to both the instrumental and the noninstrumental models of change strategy. That is, which barriers are given greater relevance by the change agent? In the case of the instru mental model, those barriers are of a cognitive nature— informational scarcity, shortage of technical skills, inadequate coordination— much like Bennis1 technocratic change, which "relies solely on collecting and interpre ting data" and in which "difficulties [derive] from inade quate knowledge." Valuational barriers are either omitted or relegated to a second order of importance, in the instrumental model, and in the technocratic change as well, for "lack of knowledge is accidental or a matter of neglect— not something that is functionally part of him 6^Ibid.. p. 155. Italics added. [the client system, which may be, of course, an individual or a bureaucracy].” Therefore, Bennis' technocratic change seems to clarify the concept of obstacle mapping, as well as the distinctions which, in regard to that concept, are established between the two models used in this study. Gouldner's Models and the Notions of Strategy Ideology and Strategy Behavior Alvin Gouldner provides good illustrations of what is here meant by strategy ideology, and of how dif ferent ideologies about change result in different strategies for changing. In a discussion of approaches to the study of complex organizations,*^ he distinguishes between the rational and the natural-system models of organizational analysis. The rational model, which is Weberian-based, conceives of the organization as an *^Alvin W. Gouldner, "Organizational Analysis," in Sociology Today— Problems and Prospects, ed. Robert K. Merton, Leonard Broom, and Leonard S. Cottrell, Jr. (New York: Basic Books, Inc., 1959), pp. 400-28. 87 instrument, and fundamentally implies a mechanical model. Clearly, the rational model requires from the client system similar or even the same authority and time patterns which are required by the instrumental model of change strategy.65 This is inferrable from the fact that: . . . The rational model assumes that decisions are made on the basis of a rational survey of the situa tion, utilizing certified knowledge, with a deliber ate orientation to an expressly codified legal apparatus.66 Since the model assumes such behaviors, a change strategy of this sort necessarily requires them from the client system if that change is to succeed. In the terminology of the instrumental model, Gouldner's words mean that the prevailing patterns of executive behavior and of expectations about such behavior are assumed to be pos it ion-oriented, and of a functional consequence for the client system. This is suggested by the assertion that behavior is oriented by, and toward, "an expressly codi fied legal apparatus" (position-orientedness) and ^Ibid.. p. 405. ^See Figure 4. 66Gouldner, on. cit.. p. 404 88 "utilizes certified knowledge" (knowledge that is func tional for the organization). Correlatively, the prevail ing temporal orientations in the client system are of an optimizing-organizational pattern. An even better example of the functional position- orientedness which is assumed to prevail in the client system may be drawn from the following excerpt: "Individual organizational elements are seen as subject to successful and planned modification, enactable by deliber ate decision.''^? for it is implicitly assumed that once a decision is taken it is implemented, because (still another implicit assumption) subordinates obey their superiors, the decision-makers. Leaving the dimension of strategy ideology and turning to strategy behavior, the rational model of organ izational analysis and the instrumental model of change strategy also have a point of great similitude to each other. Their focus of action is the same, namely, formal elements; this commonness is so explicit that it dispenses with further comments: 67ibid., p. 405. Italics added. 89 The focus is, therefore, on the legally prescribed structures— i.e., the formally "blueprinted" patterns— since these are more largely subject to deliberate inspection and rational m a n i p u l a t i o n .68 Contrariwise, in the natural-system model, the focus of analysis is on . . patterns of belief and interaction that arise even within the rationally planned organization."69 Therefore, the natural-system model has the same focus of action of the noninstrumental model of change strategy, informal elements. Still another comparison may be drawn between Gouldner's models and the ones used in this study. As he suggests, departures from rationality, in the rational model, are assumed to be derived "... from random mis takes, due to ignorance or error in calculation."^ That is, the obstacles to be removed in the change process are of a cognitive, not of a valuational, nature. This is exactly the case of the obstacle-mapping process of the instrumental model. By contrast, the natural-system model 68lbid.. pp. 404-405. 69Ibid.. p. 407. 7 0 M&-, P. 405. 90 considers departures from rationality . not so much as due to ignorance or error, but as arising from con straints imposed by the existent social structure."71 The natural-system model is, therefore, valuationally, rather than cognitively, oriented, a characteristic which is also found in the noninstrumental model. In an earlier article, Gouldner had already made a distinction between an engineering and a clinical approach to c o n s u l t i n g . Such a distinction seems to have been the basis for his rational and natural-system models, which have just been discussed. Thus, the engineering model, like the instrumental model, is cogni tively oriented, as far as its obstacle-mapping, process- is concerned. The engineering sociologist "... defines his difficulties as deriving from inadequate knowledge. He formally operates on the assumption that if he only had 73 greater knowledge, his problems would capitulate. ?- IIbid., p. 406. Italics added. ^Alvin W. Gouldner, "Explorations in Applied Social Science," Social Problems. Ill, No. 3 (January, 1956), 169-81. 73 l b i d . . p. 177. 91 Also, ". . . he {the social engineer) is concerned, for example, about problems of sampling, questionnaire design, or interviewing technology largely as these affect his data collection from respondents."74 Finally, . . he [the social engineer] assumes that his relationship with his client is regulated by the postulate that ignorance is evil, and knowledge power, and that men unequivocally prefer enlightenment to ignorance."75 xt becomes then clear from those quotations that the engineering model is akin to the instrumental model, whereas the clinical model shares some of the orientations of the noninstrumental model. In the former two, the change agent assumes that the major obstacle to change is in the client system not knowing how or what to change, whereas in the latter two, the main barrier is in the client system not wanting to change. The first is a cognitive obstacle, the latter a valuational one. 74Ibid.. p. 179 75Ibid. 92 Reiman* s Typology and the Notion of Language Stvle In an effort to identify qualitative differences between modes of social influence, Herbert C. Reiman arrived at a threefold typology.^ His processes of social influence are compliance, identification, and internaliza tion, each of them having a specific set of antecedent and consequent conditions. Actually, Reiman*s models are much less comparable to the ones used in this study than the models developed by Chin, Bennls, and Gouldner, which were examined above. However, some comparability exists in regard to the prescriptiveness and suggestiveness that distinguish between the language styles of the instrumental and noninstrumental models. The portion of Reiman* s 77 paradigm which is of relevance here is presented below: ^Herbert C. Reiman, "Processes of Opinion Change,” The Public Opinion Quarterly. XXV, No. 1 (Spring, 1961), 57-81. 7?Ibid.. p. 67. Italics in the original. 93 Compliance Identification Manner of Limitation Delineation of achieving of choice role require- prepotency of behavior ments the induced respjpioge Therefore, whenever change is approached through the presentation of "ready-to-go" solutions, or through the elimination of alternatives which might emerge from a heuristic orientation, a compliance process takes place. The compliance model is then predominantly mandatory, a characteristic which is shared by the instrumental model, whose language style Is prescriptive. Here, solutions are "musts" to be pursued, little or no discretion being left for alternative courses of action, should anticipated 78 conditions evolve. Contrariwise, when "... the induced response becomes prepotent because there has been a reorganization 7®See the operational definition of "Prescriptive,1 , Chapter I, pp. 42-43. Internaliza tion Reorganiza- of means-ends framework 94 In the individual's conception of means-ends relationships influence will tend to take the form of internalization."?^ This characteristic of the internalization model implies a suggestive language style like that of the noninstrumental model, in which solutions are "mights" subjected to modi fications according to the circumstances. The internali zation model also implies a concern for values as the pri mary blockage to change, rather than the cognitive orienta tion of the instrumental model. Another aspect of Kelman's constructs may be helpful in explaining which consequences are likely to result from a compliance (instrumental) or from an inter nalization (noninstrumental) process. This point is not considered in the models used in this study, because of the methodological difficulties of probing into it in a satisfactory way. Kelman distinguishes between the condi tions of performance of induced response in the three processes:®0 ?^Relman, op. cit.. p. 69. 80Ibid.. p. 67. 95 Internaliza- Compliance Identification tion Conditions of Surveillance Salience of Relevance performance by influenc- relationship of value to of induced ing agent to agent issues response Therefore, the compliance model requires a perma nent control by the change agent in order to maintain the intended change, which is not the case in the internaliza tion model. As will be shown in the next chapter, the need for, and the actual exercise of, a permanent control was a major characteristic of most of the Brazilian administrative reforms. This is especially true in regard to those reforms which were instituted during dictatorial periods. Furthermore, once the controlling power of the reformers was reduced by the political redemoc.ratization of the country, the reform effort received a setback which has not been halted yet, Most of the achievements made under an instrumental strategy of change entered, then, a progressive stage of dilution. The instrumental model did not provide for the internalization of the values 96 which would maintain the reform work. Therefore, such a distinction between the compliance and the internalization models may come to be useful in explaining how and why the administrative reform movement in Brazil had its greatest accomplishments under totalitarian conditions and its most serious setbacks after the reinstallation of a democratic regime. Jones and Niaz's Classification of Strategies Garth N. Jones and Aslam Niaz®*- resorted to Etzionifs compliance-power typology— coercive, normative, and utilitarian— and elaborated a tenfold classification of change strategies. Each of these strategies has a dis tinct set of t a c t i c s . Coercive strategies, for example, ®*-Garth N. Jones and Aslam Niaz, "Planned Organi zational Change: A Classification Scheme for Analysis," in Planned Organizational Change: A Set of Working Docu ments . ed. Garth N. Jones (Los Angeles: Center for Research in Public Organization, School of Public Admin istration, University of Southern California, 1964), pp. 1-39. ®2"In general terms, strategy refers to the plan ning and directing of operations; while tactics relate to the maneuvering of forces into position of advantage." Jones and Niaz, ibid.. p. 28. The distinction is far from operational as the authors themselves recognize: "It is 97 utilize four different tactics: (1) pressure tactic, (2) stress induction tactic, (3) hierarchy tactic, and (4) elite corps involvement tactic.^ Actually, there is no operational difference between a pressure tactic and a stress induction tactic, nor between a hierarchy tactic and an elite involvement tactic. Nevertheless, the hier archy tactic has the same focus of action of the instrumen- model, namely, formal elements, for it ", . . emphasizes superior and subordinate relationships between individuals never really possible to say where tactics leave off and strategy begins, but the distinction does exist between day-to-day operations and broad policy objectives." Ibid., p. 29. ^^In subsequent writings, Jones and Niaz tried to refine their classification. For example, they have re placed "strategy of pressure" for "pressure tactic," and given tactics a treatment of their own. See Garth N. Jones and Aslam Niaz, "Strategies and Tactics of Planned Organizational Change: A Scheme of Working Concepts," Philippine Journal of Public Administration. VII, No. 4 (October, 1963), 275-85. The changes, however, are not significant at all. Even in his last article on the sub ject, Jones asserts: In a narrower sense, agents of change employ what is termed here as tactics to facilitate change. This is the maneuvering of forces of various types into posi tions of advantage. Frequently, this is accomplished in connection with a strategy. In other cases, tac tics are used by themselves and may be regarded as strategies. Garth N. Jones, "Strategies and Tactics of Planned Organi zational Change," Human Organization. XXIV, No. 3 (Fall, 1965), 199. The distinction between strategies and tactics 98 in the organization."8^ Also, in such a tactic, "... management control is exercised through formal structure that relates to a system of individual subordinates with superiors."^5 The analysis of Jones and Niaz's contribution is a must for this survey, for they have the merit of addres sing themselves specifically to strategies of change. However, among the contributions reviewed in this section, theirs is one of the least directly related to the models used in the present study. This is so mainly because Jones and Niaz confined their classification to one refer ent, power. Therefore, their strategies are modeled after the mutualness of goal-setting, or the manipulation of symbolic, or material, rewards, as the major source of power in the change process. Since the instrumental and noninstrumental models are built around other (and, incidentally, much broader) referents, such as action, content, and value, the utility of Jones and Niaz's con- is still unsatisfactory. Jones and Niaz, "Planned Organizational Change: A Classification Scheme for Analysis," p. 30. 85Ibid. 99 structs for the clarification of those models is greatly reduced. Should power be included as one of the compon ents of the instrumental and noninstrumental models, then Jones and Niaz's strategies might be of much value here. Two Theoretical-Empirical Works When one turns to empirical works, these often lack an explicit model of intervention. This difficulty was already met by Lippitt and his colleagues in their collection of case studies which might be grouped together as an application of a given approach to change.88 Such a . , split between those who act and those who con ceptualize . , .,l87 seems to explain the inability of many change agents to deal in conceptual terms with their diagnostic orientations and change methods. Indeed, pieces in which the concerns for intervention and concep tualization coexist are rare. Two well known exceptions 88Lippitt, Watson, and Westley, op. cit. 87Ibid., p. 17. 100 go are the ■works of Paul R. Lawrence, at the organizational level, and of Allan R. Holmberg,®^ at the community level. Because of their unique combination of theory with prac tice, and of their usefulness for exemplifying aspects of the instrumental and noninstrumental models, those two studies are briefly reviewed here. Lawrence’s study of Food World Company. The con tribution by Paul R. Lawrence focuses on a medium-sized supermarket chain, Food World Company, the top management of which attempted to implement a decentralization. The author did not act as a change agent, but only as a researcher who followed the whole process of change in order to theorize about it. The top management of the chain was the agent of change. In spite of this split ®^Paul R. Lawrence, The Changing of Organizational Behavior Patterns— A Case Study of Decentralization (Boston: Harvard University, Graduate School of Business Administration, Division of Research, 1958). 8^We are referring to the Vicos case, in Peru. There are many reports and articles on this experience. The source from which most of this review is drawn is: Allan R. Holmberg, "The Changing Values and Institutions of Vicos in the Context of National Development," The American Behavioral Scientist. VIII, No. 7 (March, 1965), 3-8. 101 between acting and conceptualizing, there was a concomit ant preoccupation with both aspects, for the top manage ment and the researcher worked side by side from the beginning of the experience. In its broadest lines, Lawrence’s design resembles the one used in this dissertation. He built his study around two major sets of questions, the first of which illustrates very clearly his concern for identifying what has been called here the ’’ behavioral profile” of the organization, as well as for the compatibility between such a profile and the change strategy to be used: "VJhat is the nature of the basic behavior patterns in this organization? What are the key factors involved in chang ing these patterns?"^® In order to answer these questions, Lawrence used the research methods of interviews and direct observation. The second major set of questions was: "Did a significant measurable change occur in the behavior patterns of people? If so, how was it accomplished?”^^ ^Lawrence, op. cit., p. 3. 9*- Ibid., p. 4. 102 In regard to the models used in this dissertation, the experience of the Food World Company seems to have been based on a change strategy which presents some features of both the instrumental and the noninstrumental models. For example, the strategy relied entirely on the chain of command, starting with a structural change in the top management and moving down the hierarchy to the store level. This characteristic, which is also a feature of the instrumental model, may be inferred from the three following excerpts: The first specific step in the plans of these executives [the change agents] was to make certain structural and personnel changes in top management that would enable these men to proceed to make changes further down the line.92 These executives believed that the most important part of their reorganization plan was the development of a new position in the organization. the store manager. They believed that if they could develop the right kind of over-all manager in each one of their larger supermarkets, they could go a long way toward achieving their general objectives.93 ^^Ibid.. p. 52. Italics added. 93Ibid.. p, 53. Italics added. 103 Closely related to the idea o£ introducing a strong store manager were certain changes in the supervisory link between the home office and the stores. The principal feature of the plan at this level was to shift the assistant district manager from the line function of directly supervising the peris- able departments in each store to the staff Job of acting as merchandisers. Where these men had been reporting to the district managers, they were now to report to their respective merchandising managers (formerly called buyers). The objective of this change was to build a single unified chain of conroand from the store operations manager through the district aangrerg, 4iregUy,-t;p- the .pgsL.gtgre mmmotc* • • Thus, the change strategy required given behavioral patterns to prevail in the client system. Lawrence's dis cussion of the case shows very well the different conse quences that result from situations where such a require ment is, or is not, met by the client system. In his sample of three pairs of district managers (DM) and store managers (SM), one pair had changed, one was changing, and the last had not changed its supervisory pattern after the change a t t e m p t , in analyzing the self-concepts of the district managers, Lawrence points out that the self- concept of the DM who best facilitated the envisaged change ", . . so neatly complemented the new role require- 9^Ibid.. p. 55. Italics added. 95Ibid., Chapters IV, V, and VI. 104 ments and he waa not put into conflict . . . ,"96 whereas the DM who had not changed experienced a conflict between his self-concept and the intended changes. Furthermore, ". . . his [he refers to the DM who had not changed!) lack of an explicit awareness of self-environment would not initially allow him to see this conflict.''^? One may hypothesize, from these observations, that when a conflict exists between the prevailing behaviors or orientations of the client system and the change strategy, the first is likely to remain unchanged. Or, in the words of Lawrence: • . . The desired changes were not going to be executed successfully until what made "sense" in the realm of organizational welfare was reconciled with what made "sense" in terms of the self-concepts of the strategic individuals.98 As noted before, the change strategy followed in the Food World Company presents features of both the instrumental and the noninstrumental models. Thus, while relying entirely on the chain of command, focusing on the formal structure, and being concerned with cognitive 96jbid., p. 172. 97Ibld.. pp. 172-73 98J!2M., p. 173. 105 obstacles to the change process,99 the strategy also took into account the patterns of interaction between, and the sentiments of, the members of the client system, thereby focussing on the informal elements of the organization and on valuational obstacles as well. This concomitance of formal and informal elements in the change strategy is made clear in the analysis by Lawrence. He adopted Homans* classification of the elements of behavior— activities, interactions, and sentiments— and described the efforts of the change agents directed to each of those elements.^0 In general terms, one might say that the instrumental part of the strategy dealt with activities (what people do), whereas the noninstrumental one centered on interactions (verbal and nonverbal contacts between people) and senti ments (what people feel and think).Finally, in his summary of the sorts of influences used to change the behavior patterns at the Food World Company, Lawrence 99gee the quotations corresponding to footnotes numbers 90, 91, and 92, above. ■^George C. Homans, The Human Group (New York: Harcourt, Brace & Co., 1950), especially Chapter II. lOlLawrence, op. cit.. p. 6. 106 stresses the importance of encompassing both formal and informal elements in the strategy of change. He actually suggests that such a combination was responsible for what- ever success the change attempt had: It was the combination of influences that deserves credit for the change. The top management people had supplemented their initial formal organizational moves by an array of continuing influences . . . including training programs especially designed for the introduction of the envisaged change. . . . The resulting changes in behavior patterns had been achieved bv the initial formal changes together with this combination of persistent and consistent daily influences.^Q2 Holmberg's study of Vicos. The Vicos case differs from the Food World Company in several respects. The former took much more time and was much more complex, for it dealt with a whole community; also, while in the latter action and conceptualization were performed by different persons, in the former, the change agent and the theoreti cian were one and the same. Actually, concern for theory building was one of the strongest motivations that led the Cornell group to become the change agents in Vicos. 1Q2Ibid.. p. 204. Italics added. 107 The change strategy used by Holtnberg and his associates presents some characteristics which may be used for the clarification of concepts on which the instrumen tal and noninstrumental models are based. In regard to change scope, for example, the Vicos project adopted a wholistic, rather than a gradualistic, strategy, thereby resembling the instrumental model in this aspect. Indeed, one of the guiding principles of the strategy used in Vicos was that ". , . an integrated or contextual approach to value-institutional development is usually more lasting than a piecemeal o n e . "^3 But this seems to have been the only point of similarity to the instrumental model. Thus, the Vicos strategy was clearly nonprescriptive, every move coming about with consent of the Vicosines, as is exempli fied by the changes in regard to education and military service: "It must be emphasized that in no case was Indian attendance at public school or adult education classes compulsory. Even army service has been voluntary. ..104 • • • 103Holmberg, op. cit., p. 5. ^^Henry F. Dobyns, "The Strategic Importance of Enlightenment and Skill for Power," The American Behavioral Scientist. VIII, No. 7 (March, 1965), 24. 108 Also, the focus of action was not on formal elements but on the orientations of the people occupying positions in that structure. As a matter of fact, one of the characteristics of the strategy was to depend not on the immediate replacement of people in key positions, but on the recruitment of such people to, and their integra tion with, the goals of change. Aware of the importance of the traditional elite, the change agents devised strategies whereby that very elite could help the intended changes, so that the replacement of the old elite by younger men was done later and '’ naturally,'' without ten sions for the system. Furthermore, the change agents not only used the traditional elite for the benefit of change, but tradi tional institutions as well, as is best exemplified by tha so-called mando, a session of all peons, straw bosses, and the patron. This sort of meeting was strategically used to serve the function of increasing the involvement of its members with the change process, as well as to develop lO^Hoijnberg, pp. cit.. p. 6. their skills in group situations. The characteristics commented upon above may be subsumed, from the perspective of this dissertation, in one— the strategy used at Vicos, instead of depending upon certain prerequisites on the part of the client system, attempted to deal with the system as it was. The change agents avoided intervening directly in any valuationally- sensitive area, so as to minimize the incompatibilities of their values with the ones of the client system. It is quite clear from the reports on the Vicos experience that the "gyroscope" of the change agents was always more geared to valuational, than to cognitive, impediments to the intended changes. Their obstacle-mapping process was, then, rather comparable to the one of the noninstrumental model. 106Ibid.. p. 7. *0?A detailed account of the valuational orienta tion of the strategy adopted by the Cornell group is given in: Allan R. Holmberg, "Changing Community Attitudes and Values in Peru," in Social Chance in Latin America Today: Its Implications for United States Policy, ed. Richard N. Adams et al. (a publication of the Council on Foreign Relations; New York: Harper & Bros., 1960), pp. 63-107. 110 It must not be overlooked, however, that a crucial condition which occurred in Vicos imposes serious qualifi cations on the assessment of the experience. Cornell University leased Vicos and its serfs for a five-year period, and thus the change agents initiated their action as the highest authorities of the client system. Such an exceptional condition made possible certain decisions that would probably not be taken by any other group in power, such as the decision to remunerate work which, until then, had been done without any compensation. The effect of this factor upon the success of the Vicos experience is imponderable; to say the least, it seems inappropriate to attribute such a success to strategic adequacy alone. Two Unpublished Papers This literature survey is closed with a review of two unpublished contributions, both dealing specifically with models of change strategy, rather than with models of change, as is the case of most writings in this field. McEachern and Tavlor*s theory and the compatibil ity between strategy ideology and behavioral profile. The Ill first contribution, by McEachern and Taylor,*0® is an attempt to lay down a general theory, or at least a frame work for such a theory, of organizational change. This is possibly the only piece of literature which puts forth a set of interrelated propositions about organizational change in entirely operational terms. The authors' basic assumption is that: • . . Social behavior of an individual is a func tion of the phenomenological state of the individual immediately preceding his behavior. This phenomenol ogical state can be described according to two kinds of information: cognitive information about the environment, and evaluative information about the social positions he recognizes as relevant to his behavior, including the position occupied by the individual himself. On the basis of such an assumption, McEachem and Taylor undertook to speculate about which conditions might ■^®Alexander W. McEachem and Edward M. Taylor, "Inducing Change with Social Research— The Experimental Introduction of Research Findings within the Profession of Probation" (unpublished draft of a research proposal, Youth Studies Center, University of Southern California, May, 1965). 109McEachem and Taylor's draft is not numbered. In order to make possible references to it, we decided to number that paper, starting with number 1 at the page that immediately follows the table of contents. The excerpt corresponding to the present footnote was quoted from p. 21 of the draft under review. be facilitative or obstructive of any change attempt, con sidering given characteristics of both the change strategy to be used and the organization to be changed. Actually, the authors relied on a stimulus-organism-response 1 IQ model. What they call "stimulus program" may be broadly taken as corresponding to what has been called "change strategy" in this dissertation; the second term of their S-O-R-model is the organization target of change, which has been named here "client system"; finally, the last term— response— constitutes the main purpose of the theory, namely, the power of anticipating possible out comes, on the basis of knowledge on some characteristics of both the change strategy and the client system. There are many more differences than similarities between this contribution and the two models used in the present study. In comparison to them, McEachem and Taylor’s contribution is much more sophisticated, stands at a much higher level of abstraction, and has a greater measurability. However, some aspects of that theory may be related to the rationale of this study, as well as to 110Ibid.. p. 22. 113 some concepts of the Instrumental and noninstrumental models of change strategy. First, the main element of McEachem and Taylor's theory is the same one on which the whole rationale of this dissertation rests— the extent of compatibility between the change strategy, on the one hand, and the behavioral profile of the client system, on the other. In the authors' words: The extent to which the current information [in the client system] contradicts or is inconsistent with the information to be transmitted in a stimulus pro gram [of a given change strategy] is the first vari able describing organizations that we will postulate as related to the amount of information transmitted. If it is not contradictory or Inconsistent, there will be a positive effect: if it is contradictory or Incon sistent. there will be a negative effect. The word "information," in the excerpt above, is meant to encompass not only cognitive but also evaluative information, so that the consistency referred to by the authors is not confined to knowledge or "factual" aspects; rather, it applies equally to values. In this regard, the authors state: ^ • ^Ibid.. p. 26. Italics added. 114 The two kinds of information may be conceptually separated by saying that cognitive information is what an individual thinks does exist, and evaluative information what he thinks ought to exist. The former subsumes all knowledge, beliefs, opinions, perceptions, observations, and the latter all values, attitudes, expectations, norms, evaluative standards. Thus the theory is concerned with a valuational compatibility between the strategy (more precisely, the strategy ideology) and the prevailing values of the client system, values which will ultimately condition and shape the behavioral profile of the client system. Here is the closest link between McEachem and Taylor's contribution and the present study. Our basic hypothesis postulates a conflict between the strategy ideology of the instrumental model (which, we also hypothesize, has prevailed in the Brazilian experience of administrative reform) and the values which shape the behavioral profile of the Brazilian bureaucracy. It must be pointed out that McEachem and Taylor do not develop a model of change strategy. However, the "if-then" type of propositions that comprise their theory directly allows hypothesizing about behaviors which, under l-^Ibid.. p. 21. Italics in the original. 115 given circumstances, may be more facilitative than others in bringing about change. An example of such propositions is: If the chief of an organization has cognitive information consistent with a stimulus program, the effect on changes in behavior will be positive; if the chief does not have cognitive or evaluative information consistent with a stimulus program, the effect on changes in behavior will be negative. J In the present stage, the theory does not differ entiate between the relative importance of congruency on cognitive and on evaluative information to the change process. In other words, both kinds of congruency are taken as having the same impact on that process. Such a distinction is the basis for one of the dimensions of the models used here, that of obstacle mapping. In the case of the instrumental model, the obstacle-mapping process is geared toward cognitive impediments, whereas in the non instrumental model, it focuses on valuational ones. The question that arises at this point is: Which situation is potentially more obstructive to change, one of congruency about cognitive information but of incongruency about 113Ibid.. p. 33. 116 evaluative information, or one of congruency about evalua tive information but incongruency about cognitive informa tion? The assumption underlying this study is that the former situation is more obstructive to change than the latter; this assumption is clearly reflected in the obstacle-mapping process of the noninstrumental model. Contrariwise, in the Brazilian experience of administra tive reform, patterned after the instrumental model, the change agents have been primarily— almost solely— concerned with a cognitive compatibility between the client system and the intended changes. For them, the bureaucracy has not changed because of lack of knowledge, above all. These are the points of comparison between the models used in this study and McEachem and Taylor's theory of organizational change. And the conclusion with Bennis. The work of Warren G. Bennis in planned organizational change is well known today, and several of his writings have been dis cussed already in this survey. We chose to wind up this ^-^The demonstration of this point is the major purpose of the next chapter. 117 chapter with his unpublished paper,for two reasons: (1) because of the contributions of that paper to the understanding of some concepts of the instrumental and noninstrumental models, and (2) because of the conclusions at which Bennis arrived in regard to the present state of theories of change. At the outset, it must be pointed out that, in this paper, Bennis does not build any models of change strategy, in the fashion of the models of this study. His discussion of the strategies used by Blake and by the New York headquarters of a large oil firm,*--^ for instance, is too broad to allow any comparison with the instrumental and noninstrumental models. Also too general (for the purposes of this survey, of course) is his description of eight types of change programs— exposition and propagation, elite corps, human relations training, staff, scholarly consultations, circulation of ideas to the elite, develop- ^-^Warren Qt Bennis, "Theory and Method in Apply ing Behavioral Science to Planned Organizational Change" (unpublished paper, Alfred P. Sloan School of Management, Massachusetts Institute of Technology, n.d.). (Mimeo graphed.) 116Ibid.. pp. 61-68. 118 mental research, and action research.? However, his critique of the strategic rationale of these change pro grams helps make explicit some of the value assumptions of the instrumental and noninstrumental models. For Bennis, there are four particularly visible flaws or biases intrinsic in the eight programs: a rationalistic, a technocratic, an individualistic, and an insight biases.Both the rationalistic and the technocratic biases rely on the same assumptions that shape the instru mental model of change strategy. Put in summary terms, those assumptions are that knowledge equals power, that a system does not change because it does not "know" how, rather than does not "want," to change, and that, as a consequence, ". . . if a program is presented, the client can carry it out with dispatch. These are exactly the *-^Ibid.. pp. 31-36. For an extensive discussion of these points, see: Warren G. Bennis, "A New Role for the Behavioral Sciences: Effecting Organizational Change.” Administrative Science Quarterly. VIII (1963), 125-65. •^^Ibid.^ pp< 36-44. The fourth bias, insight, is not relevant to our discussion. 119Ibid., p. 37 119 kinds of assumptions which cause the obstacle-mapping process of the instrumental model to concentrate upon cognitive obstacles, with the relegation of valuational ones to a second place. Consequently, the focus of action, in this strategy, is on formal elements (systems of data processing, coordination, position classification, and so on). It is worth quoting Bennis* comments on the rationalistic bias: I do not want to minimize the importance of facts and knowledge— certainly, they do play a role in social change; yet in most of the strategies discus sed, there is an almost total reliance on rationality. Knowledge about something does not lead automatically to intelligent action, no matter how "right" the idea. Intelligent action requires commitment and programs as well as truth.120 It must be pointed out, en passant, that the inclusion of training programs in a change strategy does not preclude such a strategy from being characterized as an instrumental one. The rationalistic bias may also permeate those programs, which would then be geared to cognitive matters only— skills, information, "how-to-do" exercises— with an unconcern for valuational ones. 120ibid.. pp. 36-37. Italics in the original. 120 Incidentally, this has been typical of Brazilian training programs in public administration. The individualistic bias refers to the assumption that . .if men in power possess the right ideas, they will act in accordance with those ideas."^21 xhis point may also be related to the instrumental model, for it relies entirely on the chain of command, or the formal power structure. As argued before, the instrumentalist assumes (or his strategy requires) certain patterns of behavior to prevail in the client system. In regard to authority, for example, executives are assumed to be functionally position-oriented.122 For Bennis, the individualistic bias means, in short, a failure to approach the organization as a system, an analytical partiality which leads to a concentration on the indivi dual and not on the organization. In fact, it is an elite strategy. The instrumental model of change strategy has a cognitive-individualistic bias to the extent that it Ibid., p. 40. *-^The concept of functional position-orientedness is extensively discussed in Chapter IV. 121 assumes that if the "right" ideas are "taught" to a group of individuals in positions of command, they will be implemented, for—-another assumption— such individuals, the executives, are functionally position-oriented. If the instrumentalist does not make the last assumption explicitly, or denies it, the point remains that his strategy (in terras of the characteristics of the instru mental model as developed in this study) requires func tional position-orientedness on the part of executives as a sine qua non condition for the change to succeed. When this behavioral pattern does not prevail in the client system, the strategy is bound to fail. The other aspect of Bennis1 paper which is of interest to this survey regards his conclusions about the present state of change theories. After a review of some authoritative opinions on the subject, Bennis asserts: . . . What I find so curious is that they [theories of change] are silent on matters of direct ing and implementing change. What I object to— and I include the "newer" theories of neo-conflict, neo functionalism, and neo-evolutionary theories— is that, they tend to identify and explain the dynamic inter actions of a system without providing a clue pertain ing to the identification of strategic leverages for alteration. They are theories suitable only for observers of social change, not theories for partici- 122 pants or practitioners of social change. They are theories of change and not theories of c h a n g i n g . Bennis is actually distinguishing between theories or models of change, on the one hand, and theories or models of change strategy, on the other. Such a distinc tion has been stressed throughout this survey. The pres ent literature does not provide models of change strategy to anyone who wants to analyze the results of a change program from the perspective of the adequacy of its strategy. The researcher has to abstract, on an ex-post facto basis, the strategy which may have patterned that program, and then proceed with the identification of the characteristics of the client system which would be a deterrent to such a strategy. This was how the instrumen tal model emerged. The literature can only contribute sparse bits of information and insights, which have to be pieced together by the model builder. As Bennis points out, the lack of a theory which deals profoundly and rigorously with change strategy is probably the explana- 123yarren g . Bennis, '’ Theory and Method in Apply ing Behavioral Science to Planned Organizational Change," p. 28. Italics in the original. 123 tion for the "theoretical orphanage" of change agents and for the failures of many change programs based on theories of social change.^4 He leaves no room for doubt about his view on the matter, when stating rather positively, for Instance, "The closest approximation to a theory of change is Argyris* systems model. And again, this is_a theory of change, not a theory of changing."125 This seems to be the point to wind up this survey of the literature. Though selective rather than exhaus tive, the survey suggests the diffuseness of foci and levels of conceptualization in the study of planned change. Most of all, it reveals the lack of well-defined sets of models of change strategies, not to speak of empirical validation of such models. Bennis* technocratic change is a type of change process, not a strategy; Gouldner's rational and natural-system models refer to orientations of the organizational analyst; with few qualifications, the same holds true for Chin's system, developmental and intersystem models of change; Gouldner's 124Ibid.. p. 30. *2^Ibid., p. 75. Italics added 124 engineering and clinical approaches to consulting imply models of strategy, but they are still very incipient in regard to the definition of their distinguishing features; Kalman's typology of processes of social influence also offers some leads to the fashioning of change strategies but no more directly than, for instance, the work of Max Weber on bureaucracy; MeEachem and Taylor1 s theory is an important step towards the model building of strategies, among other reasons, because the theory is laid in opera tional terms; however, this contribution deals with the client system in terms of variables other than those focused upon by the models of this study. The fact remains that when one comes to building models of change strategies, the limitations of the liter ature are multiplied both quantitatively and qualitatively. The task of operationalizing strategies involves tremendous difficulties. This survey of the literature had two objectives. One of them was to reveal such limitations and difficulties; and another was to establish possible links between the instrumental and noninstrumental models and other studies cf planned change, so as to reinforce 125 the theoretical foundations of those models and clarify some of their dimensions. It is hoped that the survey has achieved, partially at least, its dual purpose. CHAPTER III THE BRAZILIAN ADMINISTRATIVE REFORMS AND THEIR STRATEGIC MODEL If a turning point for the advent of the Adminis trative State* in Brazil had to be demarcated, this would certainly be the decade of the thirties. The Revolution of 1930 was led by a rare species of political man— Getuiio Vargas— whose care for the purely administrative aspects of government was another one of several charac teristics which made him a unique figure in the Brazilian political history. It was under his sponsorship that most of the story to be told here was unfolded. Actually, the *We refer here to such characteristics as the growth of the civil service and the nature of the public function in modern government, as discussed by Fritz Morstein Marx, The Administrative State (2d ed.; Chicago: The University of Chicago Press, 1961), especially Chapter I. See, also: Dwight Waldo, The Administrative State— A.Study of the Political Theory of American Public Administration (New York: The Ronald Press Co., 1948), Chapters I-III. 126 127 origins of every one of the reforms discussed below are traceable to Vargas' days as the Chief Executive of Brazil, Before turning to the specific reforms, some general comments about the reform movement as a whole seer^ pertinent. This may well serve as an introduction to the more detailed analysis that follows, by outlining the distinctive characteristics of that movement in Brazil. The reforms were brought about in the wake of the 1930 Revolution. Different from the American experience, in which the symbolism of the reforms rested first, and for a long time, on the appeal of morality, later replaced by economy and efficiency,^ the symbolism of the Brazil ian reforms was a complex of morality, economy, and efficiency. The first embodied one of the most signifi cant appeals of the new regime; much potential political support from the masses might be transformed into osten- sive, overt support to Vargas if he were able to create an image of morality in public affairs. There is no doubt that the Dictator capitalized much on the appeal of o See, on this point, Waldo, ibid., Chapter X. 128 governmental morality. The second symbol, economy, was a condition for the survival of the regime itself, at least in the long run. It must not be forgotten that the 1929 depression in the United States had profound consequences for the Brazilian economy. Therefore, a most appealing message of the Revolution was to restore the economic equilibrium of the country. Finally, for the third symbol, efficiency, there are several explanations. The most important of these, perhaps, is that concern for administrative efficiency was an essential characteristic of the statesman, Getulio Vargas. His devotion to, care for, and support of, administrative efficiency was not a temporary feeling, only important in order to strengthen his dictatorship. Much to the contrary, Vargas continued to be an administrative reformist after he occupied the Presidency under democratic controls. The importance of efficiency as a reform symbol also derived from the in fluence of a group of Vargas’ advisors, notably Luiz Simoes Lopes, Mauricio Nabuco, and Moacyr Ribeiro Briggs. One of the most relevant characteristics of the Brazilian reform movement is that it was originated and gained momentum under a dictatorship. The imp locations 129 of this fact for the dynamics and fate of the reforms can not be overlooked. For example, the change process could resort— as it actually did— to a compliance, rather than an internalization, model of influence process.^ A related but different point is the public image of the main reforming agency, the Administrative Department of the Public Service (hereafter called DASP), as a creature of the dictatorship. The severe blows which almost ex tinguished the DASP after the redemocratization of the country in 1945 may be partially accounted for by that image. Another feature of the reform movement was its elitism. Indeed, a handful of experts, capable and devoted men, attempted to accomplish the gigantic task of reforming the public bureaucracy of a country like Brazil, Among the consequences of this and other factors was a centralization-and-control strategy,^ the motivational ....— ... — I ' g " k . ' ’ The distinction between the internalization and compliance models of influence used here was developed by Kelman. See Chapter 11, pp. 91-96. ^This expression was inspired in Gilbert B. Siegel's excellent analysis of the reform movement in Brazil. Centralization and control are the two variables most explored in his work. Though using a different 130 basis of 'which may be seriously challenged. A final introductory remark is worthwhile. One o? the notions of administrative theory that most influenced, and still does, the reformers is Willoughby's distinction between functional (ends) and institutional (means) or housekeeping activities.5 Several different ideas are rooted in such a distinction. Two, however, are particu- larly relevant for the strategy used in the Brazilian experience. The first argues that only by possessing efficient administrative means can a government accomplish its purposes. Therefore, the aspect to be concentrated upon is the means, or the housekeeping activities, since they are a basic condition for governmental efficiency. Among the possible consequences that may have resulted terminology and a somehow different focus, Siegel's study is comparable to ours in that, in many cases, similar criticisms are made, in different form, to the Brazilian reform attempts. Actually, this chapter draws extensively on his study. Gilbert B. Siegel, "The Vicissitudes of Governmental Reform in Brazil: A Study of the DASP" (un published doctoral dissertation, University of Pittsburgh, 1964). For a condensation of the above work, see: Gilbert B. Siegel, "The Strategy of Public Administration Reform: The Case of Brazil" (unpublished paper, University of Southern California, Los Angeles, 1965), 29 pp. F. Willoughby, Principles of Public Adminis tration (Baltimore: John Hopkins Press, 1927). 131 from this stress on the housekeeping activities was a dis regard for the actual goals to be pursued by the bureau cracy, a tendency which may be inferred from the concen tration of the reforms on personnel, materiel, and similar "adjective" areas, little being done about reforming "sub stantive" activities such as transportation and agricul ture. While it is impossible to establish any causal relationship between such a process-orientatlon of the reforms, on the one hand, and the lack of a goal-orienta- tion that characterizes much of the behavior of the Brazilian bureaucrat, on the other, one might venture to hypothesize that the process-orientation of the reformers may have reinforced— thereby becoming an agent of the status quo, rather than of change— cultural patterns which tend to confer a higher place to instrumental, rather than terminal, values. The second idea springing from Willoughby's notion of housekeeping activities was that, since those activities are common to all governmental agencies, reasons of a technical nature, and of economy and efficiency as well, dictate the centralization of those functions in a single specialized agency, to report directly to the Chief 132 Executive.6 A direct consequence of this was the concep tion of the DASP, and the reformers1 attempt to bring all the housekeeping activities under the responsibility of the DASP. A long struggle with the Ministry of Finance around the formulation of the budget proposal, of disrup ting consequences for the DASP, is only one example of the implications of Willoughby's idea. Though it is impossible to measure the extent of their influence on the shaping of the strategic model of the Brazilian reforms, it cannot be denied that the char acteristics above coranented upon did condition that strategy. Indeed, they bore considerable influence on the prescriptiveness of the language style of the reforms, on the wholism of their change scope, on the cognitive orientation of their obstacle-mapping process, on the structural focus of their action, and, most important, on the behavioral patterns that such a strategy required from the Brazilian bureaucracy if the reforms were to succeed. This chapter has two interrelated purposes. The first, and most important, is to demonstrate that the Brazilian experience of administrative reform has been 6Ibid.. p. 106 133 patterned after the instrumental model of change strategy. The other is to develop some evidence in support of the ■ J assumption made earlier that the reforms have not suc ceeded, in that a wide gap exists between their proposed goals and their achievements. Methodologically, the approach adopted was to develop a brief historical study of every reform, in the course of which its aims, results, cycle of time, and strategic characteristics are analyzed. In order to facilitate their discussion, the reforms are classified as follows: (1) the Purchasing and Materiel Reforms, (2) the Position Classification Reforms, (3) the Merit System Reforms, (4) the Budgetary Reforms, and (5) the Structural-Functional Reforms. As stated before, this is an arbitrary classification for analytical purposes only. The reforms are not independent from, but inter mingle with, each other. The identification of the strategic characteris tics of every reform rested on official and semi-official publications, as well as articles, monographs, and one ^See Chapter I, pp. 22-23. 134 doctoral dissertation® as sources of data. The scarcity of available materials and the impossibility of using questionnaires or interviews to supplement the data imposed some restrictions on the analyses of the reforms. Thus, for instance, it was sometimes necessary to resort to deduction, instead of direct reference to the content of the literature used, in order to ascertain whether the language style of a given reform was prescriptive or sug gestive. The idea of measurement, therefore, is to be taken under several qualifications in this chapter. On the other hand, the need for keeping the analyses under manageable conditions and for limiting them to the basic, more constant and general characteristics of the strategy of the reforms required the author to be highly selective in his discussion of each reform. Indeed, it would be possible to write a monograph or even a book on each of the five classes of reforms analyzed. This is not, how ever, the purpose of the present chapter. A step-by-step description of a given reform is not sought here. Rather, the effort is one of abstraction, of search for strategic ®Siegel, "The Vicissitudes . . . ." 135 similarity among the reforms. Historical recapitulation is a means to the goal of this chapter, not the goal itself. The Purchasing and Materiel Reforms As with any movement of its sort, the Brazilian Revolution of 1930 had to transform the old order, at least in part, to survive. Two of the most typical mes sages of the new regime concerned morality in government, and administrative economy and efficiency. These values were first instilled in the Brazilian bureaucracy, as a systematic movement, immediately after the Revolution of 1930. They are a by-product of that Revolution. The first reforms were directed at the activities of purchasing and materiel administration.^ The legacy o£ ^For detailed accounts of the story of these re forms, consult E. L. Berlinck, "A Reforma dos Servi9os de Material." Revista do Servico Publico. Ill (July, 1940), 127-30; Oscar Victorino Moreira, "Sistema de Material no Servico Publico Federal," Revista do Servico Publico. II (June, 1951), 36-42; and Jose""Saldanha da Gama e Silva and Daisy Florie Passarinho Pereira, "0 Sistema de Orgaos Auxiliares e de Staff na Administrate Brasileira," Semindrio Intemacional S6bre Oreanizacao. Direto e Funcionamento dos Servicos Auxiliares e Oreaos de Estado- Maior (Staff). Serie C, Tema 1 (Rio de Janeiro: Funda^ab GetifLio Vargas and UNESCO, 1952). 136 the Old Republic in regard to such activities was corrup tion, lack of uniformity due to complete decentralization, red tape, inefficiency in the acquisition and distribution of supplies and equipment, and price anarchy. These symp toms were interrelated with each other so as to form a sort of vicious circle. Since each agency procured its own supplies and equipment, it was impossible to disci pline bidding procedures or to demand quality standards, with the consequence that prices often depended more on the interpersonal relationship between bureaucrats and businessmen than on any other factor. "Also, the re quirements of the Public Accounting Code tended to restrict the field of vendors to a small group of business men who specialized in selling to the government. In order to remedy the situation, the Provisional Government decided to set up specification standards for materiel, so that the purchasing agencies could have technical references to guide their acquisitions. On the other hand, those standards would provide a basis for com paring the prices of different bidders. Thus, still in 1930, the Commission on Standards was created; it was to ^Siegel, "The Vicissitudes . . . pp. 60-61 137 be within the structure of the Ministry of Finance and 11 equipped with testing laboratories. This first attack on the problem was a complete failure. The Commission, several members of which were powerful businessmen, never functioned. In the words of Siegel, "These men failed to assemble as a group, however, and the Commission died before getting s t a r t e d . " ^ The least one can say in diagnosing this case is that the logic and rationality of the measure taken did not assure the actual commitment of both change agents and client system. Mew steps were then taken to develop a system of materiel in the public service. Thus, in 1931, the Pro visional Government centralized the purchasing activities under the newly created Central Purchasing C o m m i s s i o n : ^ . . . With the responsibility for buying and dis tributing all goods necessary to the public services. besides promoting market surveys, examining the nature, quality, and resistance of materiels to be ^-Decree 19512, Collec3b de Leis do Brasil (abridged hereafter as Coll.), December 20, 1930. ^Siegel, "The Vicissitudes . . . p. 61. ^Decree 19587, Coll.. January 14, 1931. 138 bought, and solving problems related to the simplifi cation, standardization, specification, technical testing, and control of the conservation and consump tion of materiel.14 Thus, instead of adopting a gradualistic approach, thereby progressively assuming the responsibilities of purchasing for the entire federal service, the Commission adopted a wholistic approach by extending its jurisdiction to all supplies and equipment and to all ministries at once. Also, besides purchasing, the Commission was charged with the responsibilities of the extinct Commis sion on Standards. There is little doubt that, as far as change strategy is concerned, this wholistic approach jeopardized the reform process: In spite of the lack of restraints on the Central Purchasing Commission, it soon became evident that the organization was incapable of performing all of its responsibilities because of workload; speed of proces sing purchases then became the criterion of action. Prices were made without price surveys; within a short time vendors became aware of this deficiency, and a. return to the pre-1930 pricing practices resulted. These practices added to the lack of understanding of the idea of central purchasing already prevalent, and considerable antagonism was generated toward the Commission. ^-5 ■^Gama e Silva and Pereira, op. cit.. pp. 30-31. Translation of the author. Italics added. ISsiegel, "The Vicissitudes . . . pp. 62-63. Italics added. 139 It seems probable from the above excerpt that the strategy resulted ultimately in a strengthening of the resistance to the intended reform, since the failures of the Commission were now being used as examples against the idea of central purchasing. Also, partially in consequence of the reaction against the Commission, the constitution of 1934 reintroduced the financial and auditing controls over the purchasing process, controls which had been sus pended during the Provisional Government for the sake of promptitude in the work of the Central Purchasing Commie- sion. This reduced still more the operational efficiency of the agency, thereby eliciting new reactions against it. In order to relieve the Central Purchasing Commis sion of the responsibility for setting specification standards, and also in view of the need for having such standards in the shortest possible time, a Permanent Com- 16 mission on Standards was created in 1935. At the time, there were about four hundred different types of envelopes and letter paper in use in the federal government. The differences were in size, format, texture, color, and ^Decree 562, Coll.. December 31, 1935. 140 emblems. The Permanent Commission on Standards was able to reduce those types to only four, with considerable benefits for the administration. This was the significant contribution of the Commission in its three years of existence. One must bear in mind that both the Central Pur chasing Commission and the Permanent Commission on Stand ards were created and operated during a dictatorial period.^ This is very relevant for the acceptance of change insofar as such acceptance need not depend as much on the internalization of the change objectives as would likely be the case in a regime of democratic guarantees. Rather, change agents can rely on a compliance model. In regard to authority relationships, this suggests that the need for functional position-orientedness is easier to fulfill, because the change agent has compliance power over the client system. The strategy employed by both commissions relied entirely on the chain of command. It required patterns of behavior, on the part of executives, that would strengthen the idea of centralized purchasing ^Though a constitution was promulgated in 1934, de facto situation was a dictatorship. 141 and earn credit for the reforms. Unless executives behaved in a functional way in regard to the strategy used, the reforms were bound to fail. In spite of their exceptional compliance power at the time, the change agents were often unable to elicit those behavioral pat terns which were so necessary to the reform. It may have been so because the kinds of values— goal-orientation, high esteem for the public function, depersonalization, et cetera— which would evoke the behavioral patterns required by the reforms did not prevail in the bureaucracy at all. The following case is a good example of the strategy's Inability to count on functional position- orientedness on the part of executives. The government often delayed payments to its suppliers; many of such pay ments would sometimes be carried over to the next fiscal year, with the consequence that their authorization would either never be done or would require much effort and patience from the contractors. Though monies are alloted by law in the budget, the actual liberation of such budget items represents a completely different process in Brazil. The political struggle over the approval of the budget does not end with the passage of that document; rather, it 142 starts all over again and often becomes more important within the Executive. The President uses his discretion to retard or expedite the releasing of monies so as to enhance his bargaining power. A whole set of criteria— of a "political rationality," not an administrative one— enters the scene. Further, monies not freed during the period for which they were voted cannot be spent during the following year. The most important consequence of these peculiarities of the governmental dynamics for the purchasing system was a shrinking of the commercial year, since purchasing orders could never be dispatched before the second quarter of the year. Thus, in order to utilize their resources fully and avoid the possibility of loosing their budgetary allotments, most government agencies devel oped the practice of attesting to the receipt of supplies which had not been delivered. The change agents were 18 mindful of this ieitinho. but recognized that to do ■^Brazilian idiom. As interpreted by an American: "Difficult-to-translate Brazilian term meaning unorthodox way out of a difficultyj close in meaning to 'Yankee in genuity,1 but different in that it refers not only to suc cessful mechanical improvisation, but also, like the French se ddbrouiller. to extricating oneself from man- and society-made trouble." Albert 0. Hirschman, Journeys 143 otherwise might strangle the whole purchasing system. They were hopeful, one may suggest, that controls would ta devised to check whether the supplies were actually de livered a posteriori. If so, the possible dysfunctions of the leitinho would be eliminated. This was not the case, however, for a great deal of corruption resulted from this maneuver.*® It is worth noticing that efficacy of the control a posteriori depended as much on functional position-orientedness of the public executives as did the benefits that might derive from the ieitinho Itself. Hie strategy required a kind of orientation that did not pre vail in the Brazilian public service. The bureaucracy did not lend the reforms the behavioral support they needed. ''Bureaucrats especially chafed under the controls on materiel allotments which precluded such traditional practices as the redirection of equipment and supply monev for personnel and other expenditures."2** Toward Progress (New York: Thq Twentieth Century Fund, 1963), p. 248, footnote a. l^Siegel, "The Vicissitudes . . . p. 64. 20Ibid.. p. 65. 144 In regard to temporal orientations, the strategy ideology of the materiel reforms also conformed to the instrumental model. The reforms sought to speed up the purchasing of supplies and their distribution to the min istries; they required the development of a sense of tim ing to buy during periods when the prices were the best for the buyer; they also required a concern for scheduling and for sequence in the workflow. In the terminology of the instrumental model, the reforms depended on a percep tion of time as a scarce commodity, and on a preoccupation with sequence and synchronization. As will be shown in Chapter VI, such temporal orientations were alien to the bureaucracy as a whole. In spite of their difficulties and shortcomings, both the Central Purchasing Commission and the Permanent Commission on Standards rendered some good services during the dictatorship. With the establishment of the Adminis trative Department of the Public Service, best known as DASP, in 1938, the Permanent Commission on Standards was abolished and replaced by a Division of Materiel, which 145 pi was part of the structure of the DASP. Two years later, in 1940, the Central Purchasing Commission was replaced by the Federal Purchasing Department.22 As far as strategic orientation is concerned, these replacements did not represent any change at all. It was merely a substitution of labels in order to heighten the organizational status of the two agencies. The approach continued the same, strategic action being focused on the formal structure only and the obstacle-mapping process being geared to cognitive obstacles. From then on, the efforts to improve the system of materiel were concentrated on questions such, as how to achieve a better coordination by subordinating agency X to agency Y, or by transferring given responsi bilities from one agency to another. Thus, in 1941, the testing laboratories which belonged to the former Central Purchasing Commission were transferred to the National Institute of T e c h n o l o g y .23 Soom afterwards, in 1943, the laboratories were withdrawn from that institute and 21-Decree-law 579. Coll.. July 30, 1938. ^Decree-law 2206, Coll.. May 20, 1940. 23Decree-law 1184, Coll. . April 1, 1941. 146 shifted to the Federal Purchasing D e p a r t m e n t .24 Also in 1943, the Council on Materiel Administration was created with the purpose of promoting better coordination among, and greater efficiency of, the agencies involved with the administration of materiel in the federal service.25 Finally, in 1945, the Division of Materiel of the DASP and the Council on Materiel Administration as well were sub ordinated to the Federal Purchasing Department, a part of the Ministry of Finance.26 in all this structural and functional reshuffling, one is able to ascertain from the description of the responsibilities of every new or transformed agency that, ultimately, the formal attribu tions of the materiel system did not change significantly. The prescriptive language of the reforms was typically couched thusly: Without harm to their attributions, the National Institute of Technology of the Ministry of Labor, Industry, and Commerce and the testing laboratories of the federal administration shall cooperate with the 2^Decree-law 5984, Coll.. November 10, 1943. 25i)ecree-law 5715, Coll.. July 31, 1943. 2<>Decree-law 8823A, Coll.. December 7, 1945. 147 agencies of materiel. . . .27 Whatever achievements the materiel reforms had made up to 1945, they began to fall apart thereafter. With the redemocratization of the country, many by-products of the former regime were rudely attacked. The DASP under went severe restrictions. Some of its functions were abolished, others were transferred to different agencies. The activities of materiel, for example, were ultimately subordinated to the Ministry of Finance. The change agents lost the compliance power that they had enjoyed. Now, after fifteen years, they would have to adjust their administrative strategies to a new political ideology, one which more than ever required an internalization model, not a compliance one. Would they be flexible enough to adjust themselves to the new order? The Brazilian experience of reform seems to pro vide a negative answer to the question. This is not, how ever, the point to be discussed here. What remains as a fact is that an over-all appraiual of the materiel reforms indicates that they failed, at least in the sense that their proposed objectives were not met. Actually, there 27pecree-law 2206, o p . cit.. Art. 12. Translation of the author. 148 has occurred a retrogression vith the loss of former achievements. For example, the central purchasing agency is progressively buying a smaller percentage of the supplies and equipment needed by the ministries. This retrogression is recognised by the change agents them selves. During the last years, however, the system has undergone a process of disintegration. Progress has stopped, the Federal Purchasing Department has lost terrain, the old practices of fragmented purchases under competition among the consumer agencies them selves have begun again. In summary, the materiel system of the federal civilian service not only has not followed the evolution of the whole to which it belongs, but also_revealed symptoms of retrogres sion.28 In the face of difficulties encountered by the reforms, one might expect the change agents to search for new strategies in their attempts to improve the federal service. This does not seem to be occurring, however. Indeed, the two most recent reform proposals in regard to materiel reveal the same philosophy of change, or, in more precise terms, the same strategy ideology. Thus, the 2®Ministro Extraordin&io para a Reforma Adminis trative, Plano de Execuc&o da Re forma (Rio de Janeiro: Grafica Editdra Livro S/A, 1963), p. 84. This document involves the most recent reform proposal in the area of materiel. Translation of the author. Italics added. 149 final report of the Commission of Administrative Studies and Projects (CEPA) stresses the need for reverting the materiel function from the Ministry of Finance to the DASP: Materiel is a physical administrative means. tangible, institutional, classified among the other physical means of action— personnel, housing, money, documentation, et cetera. Its management is a specific attribution of the Central Executive Agency; consequently, of the Presidency. The responsibility for studies on materiel, technological specifications, and establishment of general norms must belong neces sarily to the DASP, as the centralizing agency of the institutional activities charged to the Presidency of the Republic.29 Finally, the last reform proposal follows the same approach, adding almost nothing to the previous one. It claims that both the Council on Materiel Administration and the Materiel Division should be brought back to the DASP, but it does not consider any behavioral difficulties to be met in the implementation of the reform. In fact, the proposal implies an assumption of behavioral conform ity to its prescriptions: "The agencies of the system of 29comiss3o de Estudos e Projetos Administrativos (abridged hereafter as CEPA), A Reforma Administrativa Brasileira. Vol. IV: Relatorio Final (Rio de Janeiro: Departamento de Imprensa Nacional, 1963), p. 153. Trans lation of the author. Italics in the original. 150 materiel shall function articulately and* harmonically, independently of administrative subordination.1 1 ^ The Position Classification Reforms The Brazilian story of position classification covers almost sixty years. Since 1907, numerous attenpts to classify civil service positions have been made. Be cause of its direct bearing upon the establishment of pay levels, this is one of the most arduous areas of adminis trative change; and reform proposals on position classifi cation are a highly sensitive political issue. In the Brazilian story of position classification, only two pro posals ever achieved legislative approval; all the others died in Congress. In 1907, federal representative Justiniano de Serpa presented a bill^1 designed to reorganize both the ^Ministro Extraordinario para a Re forma Adminis- trativa, "Anteprojeto de Lei e Exposi^ao de Motivos— Reorganize o Sistema de Material do Servi$o Publico Civil, e Da Outras ProvidSncias" (unpublished material, Rio de Janeiro, July, 1953), p. 10. (Mimeographed.) Translation of the author. ^Brazil, House of Representatives, Bill No. 320 ( 1907) . 151 structure and the personnel of the federal service. According to this proposal, all federal agencies, except the Federal Treasury and the Court of Accounts, would be classified into four categories and all personnel would be classified into four ranks, somewhat similar to the British administrative class. This bill died in the legislature. Four years later, in 1911, another attempt was made. Federal representative Mauricio Graco Cardoso pre sented a bill^ aimed at distributing all federal person nel into several distinct corps, roughly comparable to careers, with a special regulation for each one. The main concern of the bill was to establish a correspondence be tween titles and duties and responsibilities among the existing positions in the federal service. At this time the incoherence and inconsistency of titles among the several ministries had led to a multiplication of salary levels, which was both the cause and effect of the absence of any government- wide systematic pay plan.^3 32 Brazil, House of Representatives, Bill No. 77 (1911). ^Gilbert B* Siegel and Kleber Nascimento, "Form alism in Brazilian Administrative Reform: The Example of Position Classification," unpublished paper, June, 1965. (Mimeographed.) 152 Even if this bill had been transformed into law, serious doubts might be raised about its implementation because, among other flaws, it did not provide any agency to be responsible for the execution of the plan. Soon afterwards, in 1914, the problem was attacked again. Federal representative Camilo de Holanda proposed several rules designed to rank the existing personnel on the basis of their pay levels. Thus, his bill stated, "Employees with equal pay shall be in the same category."34 From a technical standpoint, this just re verses the relationship between pay and either the individual’s qualifications (British practice) or the job duties (American practice). Indeed, at least in theory, pay is the dependent variable for which the independent one is either the individual or the job or both. Accord ing to Holanda's bill, however, pay levels would govern the classification system. As its predecessors, this bill never became law. ^Brazil, House of Representatives, Bill No. 212 (1914), Art. 23. Translation of the author. 153 The story is somehow repetitive throughout the twenties. The issue o£ position classification was only raised in circumstances of crisis— that is to say, inter mittently— in connection with general salary increases demanded by civil servants. On the other hand, the pro posed bills covered all personnel of the federal service. Such wholism certainly complicated political acceptance. Furthermore, even if transformed into laws, the bureau cracy did not have the administrative capabilities de manded for their implementation; the probable consequence would be a wide gap between law and behavior, "ought" and "is"— a formalistic situation. By 1929, the idea of equal pay for equal work, which is still the cornerstone of the philosophy of posi tion classification in Brazil, had already made its way into a small circle of reformers. At that time, the salary problem had reached a chaotic point. In his annual address to the legislature, President Washington Luis Pereira de Sousa expressed much concern for the problem and reiterated the principle of equal pay for equal work that should preside over the classification: 154 A careful examination of all positions in all agencies reveals that Executive Order 18588, of January 28, 1929, which was to implement Law 5662 [this law, approved in 1928, gave a general salary increase of 100 per cent to the civil service, over the 1914 levels) only acknowledged five positions on the menial staff of the secretaries of state—-care- taker, caretaker helper, office boy, mail man or messenger, and servants— as having equal characteris tics for purposes of equality of pay. . . . In all other positions, including positions of the menial staffs of all other government bureaus, there were always differences of position, place, and speciali zation which precluded an equality of duties and salaries.35 In spite of having become a matter of Presidential concern, position classification won its first legislative breakthrough in 1936 under Vargas. Again, as a result of pressures for salary increases, a so-called "Commission of Readjustment" was created and charged with the respon sibility for presenting some sort of classification plan for the existing positions of the federal service. The work of the commission was transformed into law^^ a few days after being submitted to Congress— a clear indication ■^Presidente da Republics, Mensagem Apresentada ao Coneresso Nacional na Abertura da Terceira Sessab da Decima-Terceira Leeislatura nelo Presidente da Republics Washington Luis Pereira de Sousa (Rio de Janeiro: Imprensa Nacional, 1929), p. 64. Translation of the author. ^Law 284, Coll.. October 28, 1936. This law is usually known as the "Law of Readjustment." 155 of Vargas' strength as well as of the change agents' com pliance power over the client system. The famous Law of Readjustment was not only a classification plan. It was also the first general law for the civil service in Brazil, creating the first cen tral personnel agency (the Federal Civil Service Council), establishing rules for promotion, discipline, retirement, and other personnel functions, and prescribing the open competitive examination as the major form of entrance to the civil service. Even though its full implementation was far from being achieved,^7 the reform embodied in the Law of Readjustment brought considerable discipline to the previous chaos of the public service. To be sure, this must be considered a successful reform in comparison to all the other efforts to improve position classification in Brazil, for none of the latter accomplished as much of their proposed objectives as did the 1936 attempt. ^For a detailed analysis of the administration of the Law of Readjustment, see: Kleber Nascimento, Classi- ficacao de Cargos no Brasil. Vol. 13 of Serie de Adminis- tracgo Comoarada (Rio de Janeiro: Instituto Brasileiro de Ci€ncias Administrativas, 1962), Chapter III. 156 However, the lack of continuity which has charac terized the administrative reform movement made for the gradual, but steady, disruption of the system introduced by Law 284. Though efficiently administered, the classi fication plan needed several alterations in view of changes which were occurring in the federal service. Since those alterations were not effected, the usefulness of the system was progressively reduced. As has been pointed out: Despite its defects, Law No. 284 was undoubtedly a great step in the direction of rationalization of the personnel system. Its progressive obsolescence and inadequacy were due much less to its content and policies than to a lack of continuity in efforts to adapt it to changing conditions of the civil service7^8 The strategy applied to the implementation of the 1936 reform followed the lines of the instrumental model. The focus of action was confined to formal elements, no attention being given, for example, to training, or to any process whereby the objectives and values of the reform might be internalized by the public personnel. Obstacle 3&Siegel and Nascimento, op. cit.. p. 19. Italics added. mapping centered on cognitive impediments to administra tive efficiency— adequacy of class specifications (Weber's "rulification"), complete distinction between a position and its eventual occupant (Weber's impersonalization), and well-defined lines of promotion across classes of posi tions, on the basis of the degree of complexity and responsibilities involved in the performance of their duties (Weber's specialization and hierarchization). The language style was prescriptive, the middle and lower echelons not having any participation in the decision process. Solutions were taken at the top, by the small elite of change agents, and simply dispatched downwards through the hierarchy. Clearly, the strategy depended on the chain of command to behave in a functional way so that the classification plan would not become an instrument for the attainment of personal privileges. In spite of its instrumental strategy, the 1936 reform came closer to its goals than any other in the field of position classification. It is not methodologi cally defensible to attribute that relative success to the compliance power of the DASP at the time. Many other factors would have to be taken into consideration, for the 158 circumstances of the moment were particularly favorable— the federal service was much less specialized than today, the number of employees was relatively small (about 60,000), the President was a man always concerned with administrative reform and efficiency, and the Congress lacked actual power to obstruct Executive proposals— circumstances that have never combined again. Neverthe less, the compliance power of the change agents, by de creasing the dependence on the actual internalization of the objectives and values of the reform by the client system, must have been a crucial factor for the success of the strategy used. This point is strengthened when one observes that once deprived of their compliance power, the reformers suffered their most burdensome defeats, to wit: With the fall of the Dictatorship in 1945, the system began going down the hill. Between 1945 and the second Presidency of Vargas, in 1950, the public service suffered much deprivation and maltreatment at the hands of free-booting politicians and their political supporters. Classification and other per sonnel controls were violated with little concern for the legal-rational norms that were implied. Employees frequently were able to better their employment situa tion through the courts of the land if political con nections failed. Powerful bureaucrats often unilater ally obtained benefits for favored employees [an illustration of dysfunctional position-orientedness]„ 159 The system lost, in a short period, what had required years to achieve,39 After this interregnum of five years, the reform efforts in position classification were reinitiated, again under Vargas as the President. However, he was no longer the Dictator. The DASP, though regaining much of its former prestige, could not prepare decree-laws as during the Estado Novo (1937-1945). In consequence, the political breakthrough became much more difficult for the reforms. The Law of Readjustment had remained in the legislature for a few days only. The current classification plan, however, required nothing less than six years to be approved. Indeed, presented to the House of Representa tives on September 6, 1954,^ the bill**1- was successively altered by many amendments and substitutes until July 12, 1960, when it became law.^2 An example of the distortion of the initial project is to be seen by the fact that 39lbid.« pp. 19-20. Italics added. ^Presidente da Republics, Mensaeem n. 398. de 6 de aeflsto de 1954. do Poder Executivo. ^Brazil, House of Representatives, Bill No. 4.844 ( 1954) . 42Law 3780, Coll.. July 12, 1960. 160 . . In total, 284 out of 377 classes were altered in respect to schedules of pay by the time the classification law was approved."43 The political conditions had changed profoundly, but the reform strategy remained about the same. The North American theory of position classification had total acceptance by the change agents. The new plan was entirely inspired by the North-American model. Classifi cation questionnaires were given to all federal personnel. However, if was impossible to compose class specifications from such questionnaires, so that the change agents resorted to dictionaries of occupational titles and other reference documents and to technicians of the United States Agency for International Development. The result ing class specifications were of questionable validity , to say the least. The plan was structured into positions, classes, series of classes, occupational groups, and 43siegel, "The Vicissitudes . . . p. 237. In Brazil, the Legislative not only establishes the number of pay levels and the compensation corresponding to each level, but also allocates classes of positions to the levels of the pay scale, thereby actually performing the position evaluation function. The Executive is legally forbidden from altering the allocation of a given class to a determined pay level. This is an exclusive competence of the legislature. 161 services. 44 It is one thing to compose a classification plan. To implement it is quite a different matter. Approved in I960, the Classification Law established a period of two years for its complete implementation. This implied a requirement from the client system in its perceptions of time, namely, some concern for scheduling, for meeting deadlines, for expeditiousness. However, as of early 1966, the process of transferring positions from the old to the new system— known as enouadramento— is still far from finished. At best, the rules for the installation of the plan have generated bewilderment, discontent, and total discredit for the idea of classification as preached by the reformers. At worst, they have been interpreted so that they give legal sanction to old personal privileges and open the way to new ones. One typical case may illus trate this point. The enauadramento is not based on the actual duties and responsibilities of the positions of the old system; rather, it relies solely on the titles of such 44por a detailed analysis of the characteristics of the current classification plan, see: Nascimento, op. cit., Chapter IV. 162 positions. This resulted in hasty alterations of the titles of positions in many agencies before those agencies sent to the DASP the inventory of their positions with the names of the respective occupants. The DASP then pro ceeded to the enquadramento on the basis of the titles presented in the inventory, thereby benefiting the occu pants of the re-labeled positions. For example, in one of the autarchies (entities that have some characteristics of public corporations }, the position of elevator-operator was hastily labeled "technician in ascension." Since the enquadramento takes into account the titles and not the actual duties and responsibilities of the positions, this maneuver caused elevator operators in that agency to be allocated [in the new plan] to a class of techni cians (for they were listed in the agency's inventory of positions as "technicians" in ascension), thereby being accorded a much higher pay level.45 Episodes of this sort suggest the dependence of the reform strategy on authority relationships character ized by functional position-orientedness. The change agent, the DASP, aloof in the middle of the civil service, short of specialists, and unable to analyze in loco the situation of every agency, has to depend on given orienta tions— depersonalization, willingness to accept responsi bility, sense of urgency, et cetera— on the part of the ^^siegel and Nascimento, op. cit.. p. 32. 163 client system. Such a dependence has become even more acute In view of another characteristic of the reform strategy, namely, its ldiolistic change scope. For example, the classification plan embraced not only the so- called direct administration (ministries and some depart ments) but was equally extended to all autarchies, almost doubling the number of positions to be transplanted to the new system. Therefore, in order to succeed, the reform required behavioral patterns consonant to its philosophy. The remaining characteristics of strategies which were used in the implementation of the current classifi cation plan were essentially the same as prior efforts. Action was confined to formal elements, and obstacle map ping has largely been of a cognitive nature. In the over coming of resistances, no attention seems to have been given to the implications of a value clash between the Strategy ideology embodied in the reforms and the orienta tions that shape the behavioral profile of the bureau cracy. And still, here may lie one explanation for the failure of the reforms, for: . . . While job duties and responsibilities are preached in theory and prescribed by law, other criteria . . . actually prevail. This dualism of 164 Oriteria can be viewed as a clash between the percep tions and values of the academician and technicians. on the one hand, and the politician and the public employee. on the other. In such a setting, rational technology represented by position classification can only produce formalism.46 The Merit System Reforms The struggle for a merit system in the Brazilian civil service was also unleashed with the Revolution of 1930. Vargas' commitment to the idea of conditioning entrance to the public service on one's qualifications was such that, in 1934, the precept of merit was inserted in the Federal Constitution.^ He reiterated that commitment by including again that same requirement in the Federal Constitution of 1937.^^ One would assume that the merit system had become valued in itself since, after the fall of Vargas' dictatorship, the Federal Constitution of 1946 reaffirmed the requirement of competitive examinations as the exclusive form of entrance to career positions in the ^6Ibid.. pp. 9-10. Italics added. ^Brazil, Federal Constitution of 1934. Arts. 168 and 170, para. 2. ^Brazil, Federal Constitution of 1937. Art. 122. sec. Ill, and Art. 136 (b). 165 civil service. Indeed, Article 186 of that Constitution. which is still in force, states: "The first entry to career positions will be through public examinations, in the form set by the law.'1 ^ This repeated reassurance of the merit principle in more than three decades of constitutions would suggest that the Brasilian bureaucracy has outstripped the days of the spoils system. However, this is not the case. On the basis of available statistics, it appears that merit sys tem reforms have achieved around one-tenth of their pro posed objectives. This does not imply any precise mathematical measure of the effectiveness of the reforms. Clearly, their influence upon the bureaucracy cannot be assessed in such terms alone. Nevertheless, a contrast, in numerical fashion, between the proposed objectives and the real achievements of the reforms brings some evidence to the contention that they have failed. As indicated before, the merit principle won con stitutional recognition in 1934, thirty years ago. This ^Brazil, Federal Constitution of 1946. Art. 186. Translation of the author. This is roughly the wording of the matter in the two preceding constitutions. 166 is exactly the most common period for retirement with all the attendant advantages. Had the constitutional command ment been complied with, all the federal personnel of today would have entered the public service by means of some kind of examination. By 1964, in ideal terms, all employees appointed in 1934 or before should be retired since they would have been in service for thirty years or more. On the other hand, all those appointed after 1934 should have been— as prescribed in the Constitution— under the merit system. Thus, if an index of the degree of observance of the merit system were developed, the figure for the Brazilian federal service would be one hundred: 100 per cent meritization. However, the actual story is quite different. Though there are no precise statistics about that percent age, reliable estimates center around 10 per cent. On the basis of an interview with a former director of the Selection and Training Division of the DASP, Siegel suggests: In reality, only about 12 per cent have been appointed in such a fashion through the examination process. Examinations have been systematically sup pressed j numerous temporary appointments have been made to the civil service, many of which eventually 167 became permanent through an act of Congress. A variety of other subterfuges has been used to subvert the system. In short, it can be said that the merit system has not worked.50 It is not the point to discuss the exact percent age of meritization. What seems difficult to deny is the fact that the reforms have not met their objectives. In deed, even if 20 per cent of the federal personnel were under the merit system today, this would mean that at such a rate of progress, some 120 years would still be neces sary for the full accomplishment of the goals of the reforms. In regard to change strategy, the merit system reforms followed the same pattern as materiel and position classification. First, the scope was wholistic, encom passing at once all the civil bureaucracy. Though pre scribing merit as a basis for entrance for the whole federal service, the Federal Constitution of 1934 failed to attribute the responsibility for implementing that principle to any existing or new agency. Needless to say, no steps were taken to fulfill the letter of the ^Gilbert B. Siegel, "Administration, Values, and the Merit System in Brazil," Public Personnel Review. XXIV (July, 1963), 163. Italics added. 168 Constitution. Again a wholistic orientation prevailed in 1936 when ". . . in the central personnel agency {newly created by the Law of Readjustment]— the Federal Civil Service Council— were centralized the examinations for all career positions."'*1 The Council, mainly because of its collegiate form and the lack of an appropriate staff, was able only to carry out twelve examinations, three in 1937 and nine in 1938, the year of its extinction.^2 These first attempts at introducing the merit system on a full scale and within the shortest possible time are good illustrations of what Hirschman has termed "the motivation-outruns-understending style of problem-solv ing," which is characterized by . . endlessly repeated calls for a full, integrated, definitive and rapid ^Beatriz M. S. Vahrlich, Administracao de Pessoal— Principios e Tecnicas (Rio de Janeiro: Funda$ao Getulio Vargas, 1964), p. 24. Translation of the author. Italics added. ^^Ministro Extraordinario para a Reforma Adminis- trativa, "Normas para PreservaQao e Revigoramento do Sistema do Me'rito," Rio de Janeiro, July, 1963, p. 4. (Mimeographed.) 169 solution of the difficulties that are being encounter ed.”^ In spite of the administrative limitations of the bureaucracy in undertaking such a frontal attack on the vices inherited from the past, change agents still in sisted on a wholistic approach. The Federal Constitution of 1937 was not an exception to that rule; neither was the DASP, organized in 1938, for its Selection and Training Division was given jurisdiction over all the positions of the federal government throughout the country.54 Finallyr all the subsequent laws until the current day have expli citly prescribed the merit system as the only form of entrance to all the existing positions, except those to be filled by political appointments.55 After a review of 53Albert 0. Hirschman, Joumevs Toward Progress (New York: The Twentieth Century Fund, Inc., 1963), p. 238. ^Ministro Extraordinario para a Reforma Admin istrative, "Normas para PreservacSb . . . ," p. 5. 55i)ecree-law 1713, Coll.. October 10, 1939, Art. 13, sec. VII, and Art. 17 (b); decree-law 1909, Coll., December 26, 1939; Federal Constitution of 1946. Art. 186; law 1584, Coll.. March 27, 1952, which extended the merit system to all autarchies; law 1711, Coll.. October 28, 1952, Art. 18; law 3780, Coll.. July 12, 1960, Arts. 53 and 55, which centralized in the DASP the execution of 170 the legislation in favor of the merit system, a recent reform group declared: From all this one concludes that it is clear, even emphatic, that the federal legislation determines the practice of public competitive examinations for all the positions encompassed by the classification plans of both the direct administration and the autarchies. Furthermore, such legislation also centralizes in the DASP the authority and responsibility for carrying out all such examinations, and, consequently, for the preparation of the lists of candidates who are eligible for appointment.56 The conclusion of the above mentioned reform group is clearly indicative of the awareness of the need for trying a different strategy of change in the struggle for the merit system: ' ‘ Evidently, it would not seem necessary to strengthen the merit system, for the existing legisla tion looks satisfactory."57 The foregoing discussion of the wholism which has characterized the change strategy of the merit system examinations for all autarchies, thereby doubling the responsibilities of the Division of Selection and Training of the DASP; and law 3807, Coll.. August 26, 1960. Sfi + Ministro Extraordinario para a Reforma Admin- istrativa, "Normas para Preservagao . , . ," p. 8. Trans lation of the author. Italics added. 57Ibid 171 reforms may seem exaggerated. Moreover, some might sug gest, the implications of wholism for the adequacy of any strategy may not be that important. After all, is it not only natural for public administrative reforas to tend toward the wholistic? A survey of the world experience of administrative reform in order to identify gradualistic strategies is both inopportune and beyond the qualifications of this author. However, it seems rather significant that the very model which inspired the experience of administrative reform in Brazil^8— the North American model— was an example of gradualism. As registered by one of the best historians of the United States civil service, Paul P. Van Riper: The act itself [Pendleton Act] placed only slight ly over 10 per cent of the positions in the federal public service— mainly clerical positions in Washing ton and in post offices and custom houses employing fifty or more persons— under the merit system to form the classified civil service.59 58For an elaboration of this point, see: Wahrlich, o p . cit.. pp. 25-30. 59paul p. Van Riper, History of the United States Civil Service (Evanston, 111.: Row, Peterson & Co., 1958), p. 105. Italics in the original. 172 The gradualism of the American reform was not con ceived by the Legislative. Rather, it was an essential part of the philosophy of change of the American reform ers. Their perceptions of the political and administra tive reality were quite different from those of their Brazilian counterparts. In his testimony before the Senate Select Committee to Examine the Various Branches of the Civil Service on January 13, 1881, Dorman B. Eaton, the main conductor of the reforms, expressed his views in regard to the comprehensiveness of the Pendleton Act: Another observation I want to make is that X think no law should be passed which would require the appli cation of this system of examinations to the whole civil service of the government at once, or even to all that part to which it is legitimately applicable, as I have defined it. It would be too large altogether. . . . We have got to create the machinery. . . . In bringing new men together and entering for the first time upon a new system, you would be utterly overslaughed and broken down if you were to be required to carry it all on at once.60 Clearly, Eaton's understanding of the problem did not outrun his great motivation to see the full and definite meritization of the American civil service. The 60u.s., Congress, Senate, Report 872. pp. 19-20, footnote 3, as quoted in Van Riper, op. cit.. p. 105, footnote 15. 173 gradual— but relatively steady— expansion of the merit system in the United States is indicated in Table 3. Another strategic characteristic of the Brazilian reformist movement is its prescriptive language style. Solutions are "musts," decided by a small elite at the top of the administrative hierarchy, and channeled down ward through the bureaucracy. There are many different evidences of such a prescriptiveness. The following statement is typical of laws and other documents concern ing reform plans: The ministries, autarchies [. . . et cetera . . .] whenever in need of personnel, and also in cases of an Initial vacancy to be filled, shall ask the DASP to indicate eligible candidates approved in examinations, their order of classification being strictly observed.61 A more relevant indication of the prescriptive style of the reforms may be found in the study of the Dasnian philosophy of administration— a philosophy of police, rather than of service— particularly during the dictatorship. The testimony of one of the leading figures of the DASP may suffice to justify our contention: "The **^Law 3780, Coll.. July 12, 1960, Art. 55. Italics added. Translation of the author. PROPORTION OF POSITIONS TABLE 3 SUBJECT TO COMPETITION IN THE FEDERAL SERVICE, 1884-1960 Year Number of Positions in Executive Civil Service3 Number of Positions in Competitive Service Percentage of Positions Subject to Examinations Total United States Population 1884 131,208 13,780 10.50 54,419,760 1900 208,000 94,893 45.62 74,689,889 1914 435,000 292,460 67.23 96,551,717 1930 580,494 462,083. 79.60 122,775,046 1940 1,014,117 726,827 7l.67b 131,669,215 1950 1,934,040 1,641,876b 84-89b 150,697,361 1955 2,371,405 l,991,506b 83.98b 164,303,000 1960 2,398,705 2,050,939b 85.50b 179,323,175 ^Includes positions in the General Accounting Office and the Government Printing Office which are actually within the legislative branch. **Does not include those positions in the excepted services which are covered by independent merit systems such as found in TVA. The actual number of positions subject to merit systems is, therefore, about 5 per cent higher. Source: Figures obtained from the Annual Reports, U.S. Civil Service Commission and the Conanission's staff; total population figures from the U.S. Bureau of the Census. 175 DASP's attitude reached the point of being considered as 'policing' by functionaries, public authorities, and can didates to the public service.” 62 in an attempt to rationalize such an orientation, the same writer leaves no doubt in regard to the prescriptiveness of the reforms: Such a Dasoian orientation may be, however, per fectly understandable and even justifiable, in view of the situation that the agency undertook to modify. The fact is that the transition from nepotism to merit can hardly be done in a lenient way. This is more so if we consider that an authoritarian regime was in force, at the time, and it is not in the ethos of authoritarian regimes to act cautiously and prudently as it is indispensable in a representative democracy.63 Still from another angle, perhaps more illustra tive than any other, may the prescriptiveness of the Brazilian reforms be approached. Taking prescriptive in the sense of mandatory and suggestive as permissive, one is able to contrast the Brazilian and the American reforms from the standpoint of their language styles. In Brazil, the laws, executive orders, or even general statements of reform policies have been mandatory; as a matter of fact, 62Wahrlich, on. cit.. p. 33. Translation of the author. ^Ibid. Translation of the author. 176 this approach begins at the level of the President. For example, according to the constitutions of 1934, 1937, and 1946, the President was left no choice as to whether to implement the merit system to the whole public service. The same holds true in regard to laws defining the re sponsibilities of reform agencies like the DASP. Con trariwise, in the United States, the Pendleton Act was, for the most part, permissive (suggestive) rather than mandatory (prescriptive). Except for the approximate 10 per cent of the positions included in the classified service, "... the remainder of the service was left unclassified, to be brought under the new regulations by Executive Order when mid if the President saw it fit."64 Because it is so revealing of the different perceptions between Brazilian and American reformers, it is worthwhile to include’ .here a final quotation from Van Riper. In his comments on the reasons for the permissive nature of the Pendleton Act, this administrative historian states: It was [as recognized by the change agents] both politically and administratively impossible in 1883 to 64Van Riper, o p . cit.. p. 105. The word "unclas sified" is in italics in the original. The other italics were added. 177 apply the merit system to the entire civil service. Administratively, the Civil Service Commission simply was not ready to do a complete job as yet. It takes time to develop examinations, and to obtain the cooperation of the departmental agencies and the gen eral public. Permissiveness also had its political advantages. The politicians were able to announce that they had accomplished the desired reform— the rest being up to the President— knowing full well that they would not be hurt through a too sudden or drastic curtailing of their patronage. If the act permitted an orderly retreat of parties from their prerogatives of plunder, it made possible as well the gradual administrative development of the merit system.65 Even as recently as 1940, the Ramspeck Act author ized, not demanded, that the President extend the coverage of the merit system to some 200,000 positions previously exempted from it.66 One might argue that the conditions were com pletely different in the case of Brazil, under a dictator ship. Notwithstanding, the fact remains that quite dif ferent consequences may have resulted from the different strategies of change used in the two countries. To point up only one example, in the United States, the spoils .Ibid. Italics added. Van Riper's analysis seems to be directly based on Dorman B. Eaton's testimony before the Senate (see quotation corresponding to footnote 60, above). 66Ibid., p. 344. 178 system became more and more a burden on the President, so that its extinction was ultimately an initiative of the Executive. In Brazil, however, even though the spoils system was first condemned by Vargas, because of the for malism that resulted from the incapacity of the adminis tration to meet the "wholistic" demands of the reform the Executive lost the authority but retained the power of free appointment of personnel. Indeed, much of his poli tical strength still derives from the sinecurism that marks the Brazilian public service. Thus, the future does not look very promising to the merit system in Brazil, at least in the way it has been conceived and pursued in the past. A discredited merit system may be worse than no system at all. To speak of the way in which the merit system has been conceived in a given country raises an intriguing question, one that may not have come to the minds of the Brazilian change agents of the thirties: Is it possible to develop different conceptions of the merit system? The answer is yes. More important, the strategy of the reform in this case depends much on the conception of what a merit system is. Should it be defined as a system whereby 179 entry to the public service is on the basis of one's ability to perform the duties of a position, as demonstra ted in an examination, the focus of the change agent’s action would be selection. However, if conceived as a system whereby the public servant is provided the neces sary training to better perform the duties assigned to him, so that his holding of a position be based upon his merit in performing its tasks, the change agent's action would be directed to training. In the first case, the candidate is the only one responsible for meeting the technical standards set by the state; in the latter, such a responsibility is shared between the candidate and the state. The Brazilian reformers adopted, without adapting, the idea of a merit system. It was almost a carbon of the American original, aimed at "closing the front door." Here, as in many other aspects of its reform experience, the Brazilian administrative system has changed mostly by responding to external appeals, to alien practices. In Riggian terminology, rather than an endogenous locus of change, the Brazilian experience of administrative reform has had an exogenous locus of change. The personnel legislation is deeply rooted in the American model, though not supported by corresponding behavioral patterns. The merit system vainly continues to be the golden objective of personnel administration. It is sought as a redemption in itself, and the effort is still concentrated on the entry process, at the expense of training and other per sonnel functions which, in view of the Brazilian specific educational conditions, are far more conducive to the actual professionalization of the public service. Mimicry is easier, but often much more costly than creativity. It seems reasonable to hypothesize that had the resources applied to selection been utilized for training— even during the dictatorship— the balance would have been more positive. The degree of professionalization would be higher and the internalization of the reform values would be deeper. A possible consequence of this would be a greater integration of the civil service as a pressure group for the merit system and a DASP less stigmatized ^Fred W. Riggs, The Ecology of Public Administra tion (New Delhi: Asia Publishing House, 1961), pp. 141- 43. 181 than the one that barely survived the redemocratization of the country. The foregoing comments lead to another character istic of the change strategy of the reforms, namely, their obstacle-mapping process. Here, again, the Brazilian experience followed the instrumental model. The concern of the change agents has been with the adequacy of the level of difficulty of the examinations (often unrelated to the actual requirements of the job), the drafting of proper recruitment bulletins, the establishment of a due process of appeals, and related matters. As may be inferred from several points developed in this discussion, valuational obstacles have not been given priority by the reformers. Hence, to a large extent, there developed the disregard (perhaps a sort of intellectual satisfaction, in some cases) of the early change agents for the actual fear that an examination held by the DASP evoked in candidates to the public service. This situation still persists, with negative consequences for the image of the DASP in the general public. The cognitive orientation of the reforms has been coupled with a strategy behavior whose focus of action is 182 on formal elements. This is clearly revealed from an analysis of the basic acts which defined the functions and activities of the former Federal Civil Service Council^® and of the D A S P .69 Even in the few cases where there seems to be some concern for the more behavioral aspects of administration, the focus is still on the formal legalistic side of the problem. For instance, one of the twelve functions legally assigned to the Personnel Development Division (now extinct) was "to study and pro pose norms for the regulation of intellectually mal adjusted personnel, a function which has not been performed as yet. Furthermore, even if it had been, who would assure that the regulation would be enforced? The Budgetary Reforms The first constitution of the newly independent Brazil assigned to the Secretary of Finance the responsi- 6®Gama e Silva and Pereira, oo. cit.. pp. 66-67. 69Ibid.. p. 69. 7Gpecree 11101, Coll., December 11, 1942, Art. 33, sec. IX. Translation of the author. 183 bility for coordinating the budget proposals of the sev eral secretaries of state (as the current ministries were termed at the time) into one document and to present it, 71 annually, to the House of Representatives. The consti tutional commandment was complied with for the first time seven years later, in 1831.^2 The second constitution, dated 1891, established for the first and only time in Brazil the practice of a legislative budget,^ which was assigned to the exclusive responsibility of the National Congress: It [the Constitution] assigned no role to the Executive branch. Indeed, a real issue arose between the two houses of the Congress. The Constitution did not assign responsibility to initiate the budget to ^Brazil, Imperial Constitution of 1824. Art. 172. 72 a Aliomar Baleeiro, Uma Introducao a Ciencia das Finances. Vol. II (2 vols.; Rio de Janeiro: Editora Forense, 1958), pp. 689-90. 73 The classical piece about the historical evolu tion of the national budget during the First Republic of Brazil is: Agenor de Roure, 0 Orcamento (Rio de Janeiro: Pimenta de Melo, 1926). For the more recent history, see: Guilherme Moojen, Orcamento Publico (Rio de Janeiro: Edi$3es Financeiras S. A., 1959); and Ar£zio de Viana, Orcmnento Brasileiro (Rio de Janeiro: Edi$oes Financeiras > 1950). 184 either body. Finally, after considerable struggle, the lower house inaugurated the budget p rocess.% But the inviability of a legislative budget was soon recognized and, though the voting of the budget was initiated by the lower house, the formulation of the bud get proposal was never withdrawn from the Ministry of Finance. Indeed, still in 1891, it was explicitly deter mined by law that such a ministry was responsible for the , . centralization and harmonization, altering or reducing, of the partial budgets of the ministries, so as to organize the national budget p r o p o s a l . ” ^ In 1922, by force of the newly enacted Code of Public Accounting, the initiative of the budget reverted to the Executive. The Comptroller General of the Repub lic, who was formally subordinated to the Minister of Finance, took on the responsibility for the elaboration of the budget proposal* ^Frank P. Sherwood, uThe Budgeting Process in the Brazilian Federal Government’1 (unpublished paper written in Brazil, Getulio Vargas Foundation, May 29, 1963), p. 8. 75Law 23, Coll.. October 30, 1891, Art. 1. Trans lation of the author. 185 These are the historical antecedents of the federal budget up to the advent of the 1930 Revolution. Though the administrative reforms started months after the ascendancy of Vargas to power, nothing was altered in regard to the budgetary process. It remained the responsibility of the Minister of Finance, as it had been for more than a century. The Federal Constitution of 1934. virtually imposed by Vargas, even reaffirmed the authority of that minister in regard to budget matters. By 1936, the administrative reforms gained more impetus with the creation of the former Federal Civil Service Council; but the budget was left untouched. The Estado Novo came into existence and with it the Federal Consti tution of 1937 which withdrew the budget formulation func tion from the Ministry of Finance and assigned it to an administrative department of the public service, to be created. This event may be considered the starting point of the budget reform movement in Brazil. From this brief historical sketch it may be observed that the budgetary function was the last of the traditional administrative areas targeted by the change agents. Indeed, much had already been done In materiel 186 (1931 onwards), personnel (1935 onwards), and position classification and organization (1936 onwards). This may indicate the greater political sensitivity that would make of the budgetary reforms a most difficult front for the change agents. Here, more than anywhere else, a tenacious power struggle had to be fought, a struggle for which the reformers were ill-armed: the politics of administration. The values which shaped the professional ethos of the early reformers left them incapable of using the tools and maneuvers of their adversaries, for their crusade was it self against the politics of administration. Though the Federal Constitution of 1937 had assigned the function of formulating the budget to the agency which would come to be the DASP and though the political regime was a dictatorship, the letter of the Constitution was disregarded. Hie dictator did not resist the pressures by the Ministry of Finance and, having to chose between the technician and the politician, decided for the latter. In order to save appearances, the formal coupetence to prepare the budget proposal went to the DASP. Thus, the decree-law that organized that agency declared: "Until the organization of the Budget Division, 187 the budget proposal will continue to be prepared by the Ministry of Finance, with the assistance of a representa tive from the DASP."76 It must be pointed out that the structure of the DASP, laid down by the same decree-law, already included the Budget Division, but without its responsibilities defined. On the other hand, the func tions of all the remaining divisions were specified in that act. In 1940, a Budget Commission was created in the Ministry of Finance, It was to be a temporary solution until the Budget Division of the DASP was organized. The fact, however, is that only in 1945— eight years after the constitution had prescribed it— was the budget formulation function actually turned over to the DASP, whose Budget Division was organized in that year.?? ^Decree-law 579, Coll.. July 30, 1938, Art. 3. ^It must be recognized that, because the Budget Commission of the Ministry of Finance happened to be head ed by one of Vargas' closest administrative advisors, Dr. Luiz Simoes Lopes, who was also the President of the DASP, this agency actually gained considerable influence in the formulation of the budget proposal after 1940. The point stressed here, however, involves the obstacles that the reformers had to overcome in their intent to take over the budgetary function and rationalize it. 188 The struggle had not come to an end, however. In the following year, in the congressional debates on the coming constitution of 1946, several alternatives were insisted upon, aimed at removing the budget function from the DASP.7® In 1950, there were two budget proposals in Brazil, one prepared by the Ministry of Finance and the other prepared by the DASP. President Dutra chose the first and sent it to the legislature. Meanwhile, de Bittencourt Sampaio, then directing the DASP, clandestinely supplied the DASP1s version of the budget to the Budget Committee. The Commit tee's proposal was a carbon copy of the document from DASP.79 More recently, and to a great extent in consequence of the increasing concern with governmental planning which may be noticed during President Kubitschek's mandate, the tendency has been to return the formulation of the budget to the Ministry of Finance. This tendency, now recognized and less attacked by the Daspians themselves, was rein forced with the preparation of President Goulart's Triennial Plan, as may be deduced from the following ^Baleeiro, op. cit.. p. 720. ^Siegel, "The Strategy . . , ," p. 16. 189 remark: In 1963, with the Governments finances having reached a critical state, it was clear that the budget was being prepared in terms of the broad policies laid down by the Minister of Finance. Indeed, it was reported in the newspapers that the 1963 budget pro posal was cleared bv Finance Minister Santiago Dantas and Planning Minister Celso Furtado before its pre sentation to President Goulart for final approval*80 The DASP's attempts to be assigned the responsi bility for the formulation of the budget proposal seem to have received a death blow in the so-called Amaral Peixoto Reform proposals of 1963. To be more precise, the Daspian reformers themselves, who architected that reform, sought to allocate the function to a proposed planning agency. Indeed, in his justification of the reform project, the Extraordinary Ministry for the Administrative Reform, Amaral Peixoto, stated: The transfer of such a function [preparation of the federal budget proposal] to the Secretary of the Planning Council is necessary. In the light of modem concepts, indisputably demonstrated by the practices of several countries, the budget must be nothing more than an annual phase, pre-accounted, of long-run ®^Sherwood, op. cit.. p. 17. Italics added. 190 governmental plans, Though the Amaral Peixoto Reform was never approved, new ones have been presented which support the conception of the budget as an instrument of planning of the responsibility of the planning agency. There seems to be no receptivity for any rationale by which the DASP would regain that responsibility. The task of pinpointing the characteristics of the instrumental model of change strategy in the Brazilian reform experience is considerably more difficult in regard to budget than to any other area. At best, the analytical approach has to be different here. To look for behavioral versus structural orientations of a reform in personnel is less puzzling than in budget. Nevertheless, it is pos sible to identify several symptoms which suggest that the strategy of change which underlied the budget reforms was roughly similar to that which guided the other change attempts. In the struggle against the Ministry of Finance, for example, one may detect a disregard by the ®^Ministro Extraordinario para a Reforma Adminis trative, Reforma Administrative Federal de 1963— Antenro- ieto de Lei Organica do Sistema Administrativo Federal. Vol. I: ExposicSo de Kotivos (Brasilia: 1963), p. L7. Translation of the author. 191 reformers (in this case, more than disregard, even rejec tion) for the valuational implications of transferring the formulation of the budget function to the central agency. The reformers knew that for more than one century the Ministry of Finance had prepared the budget proposal. They also anticipated the tremendous resistances to be elicited by any strategy which required the withdrawal of that function from the finance agency. However, since it was more "rational," for reasons of efficiency, economy, and technical excellency to group all the housekeeping activities under a central staff agency directly under the President, "as Willoughby taught,"®^ the reformers decided for this solution.Had their obstacle-mapping process been valuationally-oriented rather than attuned only to cognitive barriers, they might have tried to improve the ® Willoughby, more than Gulick or any other theorist, influenced the general philosophy of the admin istrative reform movement in Brazil. Many pieces of the Brazilian literature show this very clearly. For a most incisive demonstration of this fact, see: Gama e Silva and Pereira, op. cit.. Chapter II. OQ As a matter of fact, this solution was never adopted in the United States itself, where the idea emerged. In this country, the budget and the personnel functions were split between two independent agencies. 192 budgetary process by means of a different strategy. After all, would it not have been strategically wiser to seek to improve the budgetary process through the Ministry of Finance instead of against it? How rational was it to elicit an obstacle that might well have been manipulated to facilitate change, if only for the sake of a complete centralization of the housekeeping activities? Is it the goals or the means that count? These questions are clearly the rationale for the following remark made by an attentive observer of the Brazilian administrative dynamics: Attempting to bring about change in cultures in which there are values that are inimical to bureau cratic norms is indeed an ambitious undertaking. While it is difficult to change culture, it is equally obvious that the instruments of change have to be selected with great care. For example, in the lone run, the Ministry of Finance— and not the DASP— might have been the most relevant element in the change game in Brazil. This is reasonable in view of the great residue of power that has traditionally accrued to that agency. Though the DASP won minor skirmishes. the Ministry proved to be the more durable institu tion.^ This same example may illustrate other elements of the strategic model of the reformers. As to strategy ®^Siegel, "The Vicissitudes . . . ," p. 322. Italics added. 193 behavior* for instance, their focus of action centered upon formal elements. The whole argument with the Ministry of Finance resulted, after all, from the change agents' concern with reshuffling the structure of the civil service, so as to integrate the housekeeping activities under a central agency. Other symptoms of this formal orientation were the stress on the legal aspects of the budget, its procedures, its accounting control, all at the expense of the basic purpose of the budget, namely, to reflect' the economic and financial policies of the government. The overcentralization and control of bath the preparation and the execution of the budget stimulated, on the one hand, ieitinhos to circum vent the norms, and, on the other, "... the frequent use of practices that, though fulfilling the accounting legal requirements, actually subverted the moral and administra tive aspects of the execution of the budget."85 It is not defensible to criticize the budget reforms for their wholism, as was done in the other B^Mhrio Wagner Vieira da Cunha, 0 Sistema Admin istrative) Brasileiro (Rio de Janeiro: Ministerio da Educa$ao e Cultura, 1963), pp. 93-94. Translation of the author. 194 fields. This is so because the assumption of gradual development of budget norms or the gradual inclusion of ministries in the national budget does not make much sense. However, if approached from another angle, the notion of wholism may be identified— and criticized as well— in the budget reform movement. Indeed, wholism may be understood as implying a short-time perspective. That is, it may represent a compulsion to achieve a full, rapid, integrated, and definite solution instead of allow ing for some sort of maturation of the change process. In this sense, the reforms were wholistic, as becomes clear in the effort to withdraw the formulation of the budget from the Ministry of Finance. The reformers did not want to wait; they did not want "half improvements," Rather, if rid of any interference from any other agency or authority, they would be able to do "what needed to be done," to take a surgical, rather than a treatment, approach to the client system. This kind of wholism was highly disruptive of the reforms and also of the DASP, as 86We are inspired here in Hirschman1s idea of a motivation-outruns-understanding approach to problem solving. See p. 168. 195 shown in the present section. Actually, most of the post* 1945 legislative proposals aimed at emasculating the DASP, either in part or in toto. are related to its long and inglorious battle around the formulation of the budgetary document. Finally* in regard to time patterns* there seems to be no need for dwelling on the point that the strategy of the reforms followed the instrumental model. The requirement of some minimum concern for scheduling* for Instance* is evident from the several deadlines determined by law in regard to the different phases of the budget. The need for scheduling is particularly pressing during the formulation of the budget proposal. Hare* also* a minimum concern for synchronization* sequence* and rate is required. This is so because the central agency depends on the inputs of administrative units throughout the public service. Such inputs must come at a given time* in a given order of arrival* and, therefore* in a given rhythm. Should it not be so— as has often been the case— the work of the central agency may be totally disturbed, with the consequence that* in order to submit the final proposal to the President within a determined deadline, 196 the DASP has to estimate (often by making extrapolations from previous budgets) the amounts as well as the destina tions of monies for the ministries. For these reasons, the reforms depended upon the temporal orientations that the instrumental model requires from the client system. The Structural-Functional Reforms This last front of the Brazilian administrative reform movement deals with those efforts which, instead of focusing upon materiel, personnel, or budgeting, were con cerned fundamentally with either the structural, the func tional, or both aspects of the governmental organization. Put in simple terms, the idea here is that more efficiency can be attained if the structural composition of the bur eaucracy tends to facilitate, rather than obstruct, the administrative process. Hence, lines of authority and responsibility are clarified, organizational units are subdivided or collapsed to attend the principle of homo geneity, or an adequate span of control is sought by the reduction in direct subordinates of the chief executive. A similar rationale underlies the diagnosis of the func tional aspect of the problem. Workflows are rectified, 197 routines are shrunk, or forms are simplified* These com prise what, for the purposes of this study, are termed structural-functional reforms. In order to facilitate their analysis, the structural-functional reforms may be roughly classified in two kinds, namely, functional and structural reforms, although they intermingle with each other* We now turn to the former. The Functional Reforms Historically, the starting point for the function al reforms--those of an 0 and M sort, with stress on methods— is the Law of Readjustment of 1936* Since then, a succession of efforts towards the rationalization of work methods and processes has taken place in the Brazil ian federal government. The famous law, which encompassed other areas as well, created the Federal Civil Service Council and prescribed the following functions for it in regard to organization and methods: a. To study the organization of the public ser vices and propose to the government whichever measure be necessary to its improvement; b. To give opinions on proposals, procedures, and 198 programs of rationalization of the public services prepared by the Efficiency Commissions.87 There was one commission in each ministry. They were created by the same law and also dealt with person nel. In respect to organization and methods, the efficiency commissions were to: a. Study permanently the organization of the services of the respective ministry for the purpose of identifying causes of reduced performance; b. Propose to the minister modifications judged necessary for the progressive rationalization of services; c. Propose alterations judged convenient in the staffing or restaffing of bureaus, services, or establishments. This system existed only in the law. It never even initiated its activities in the 0 and M area. Indeed, both the Council and the efficiency commissions were entirely consummed with personnel functions. Nor in personnel was much accomplished. As shown before, the Council carried out twelve examinations in its two years of existence.*^ jn 1938, the DASP came into being. ®^Law 284, Coll.. October 28, 1936, Art. 9 (a and b). Translation of the author. ®®JCbid., Art. 17 (a, b, and c). Translation of the author. ®^See p. 168. 199 The decree-law which organized the DASP created in its structure a Division of Organization and Coordination. The former efficiency commissions, composed of five mem bers, were abolished and new ones created with three members. Contrary to the practice which prevailed during the Federal Civil Service Council's period, the members of the new efficiency commissions were relieved of their normal activities so as to devote full time to the new task. However, in spite of the total failure of the pre vious system due to the combination of the personnel and organization functions, no change was made in this regard.^® The consequence of such a combination— again a symptom of wholism— was that the new efficiency commis sions were as ineffective in their organization and methods obligations as their predecessors. As recognized by two early reformers: In spite of that measure requiring full time of their members, the commissions were not able to sat isfy their objectives, not even having started, for a certain time, the activities which constituted their main purpose: the rationalization of ministerial 90oecree-law 579, Coll.. July 30, 1938. 200 services through the permanent study of their organi zation.^1 The same authors justify such a failure: The reason is that the commissions were overtaken by the excessive workload regarding personnel prob lems, such as: preparation of promotion proposals, inspection of admission processes, and reintegration and melioration of extranumeraries. Besides, the practice was established of submitting to the commis sions other matters entirely strange to their pur poses, whenever such matters involved some complex- Actually, the efficiency commissions initiated their activities in the area of organization and methods only in the second half of 1941, when all the personnel functions were withdrawn from their jurisdiction.93 How ever, in the same year, the efficiency commissions of the 94 military ministries were extinguished. And three years later, the system started to fall apart. Indeed, with the redemocratization of the country in 1945, the DASP was ^Gama e Silva and Pereira, on. cit.. p. 47. Translation of the author. 9^Ibid. Translation of the author. ^Decree-law 3569, Diario Oficial. August 29, 1941, Arts. 1 and 2. 94uecree-law 3838, Coll.. November 19, 1941. 201 almost emasculated. Thus, the same decree-law which virtually destroyed that agency^ also affected the 0 and M efforts by bringing together in one unit--the Division of Budget and Organization— the formerly distinct Division of Budget and Division of Organization and Coordination. The final blow was not long delayed. About six months later, all the efficiency commissions were abolished by a new decree-law.^6 Though it might appear that President Dutra had extinguished a fully-operating system, this was not the case at all. As Siegel points out: At the time of their elimination, however, the efficiency commissions were virtually vestigial organs— for the most part not functioning and in many cases without personnel.^ This was not the reformers' perception of the problem. For them: This extinction was a deadly blow to the system. The Council of Administration [DASP] did not meet any more to examine problems of organization, and the Division of Organization no more functioned in arti culation with ministerial units of organization, ^^Decree-law, 8323-A, Coll,. December 7, 1945. ^Decree-law 9503, Diario Oficial. July 23, 1945. ^Siegel, "The Vicissitudes . . . ," p. 206. 202 limiting itself to act only if asked for bv whichever units might be interested in the reform.*” The italics above, added by this author, suggest how much the reforms depended on given behavioral patterns to prevail in the bureaucracy. To the extent that values such as rationality, efficiency, and economy had not been internalized in the public service by the fall of the dictatorship, most of the possible accomplishments of the reformers during their technocracy might not survive the return to democracy. The compliance model was useless now. Since the Division of Organization could only act if solicited by the field units, the reforms required those units to be interested, motivated, and cooperative. In terms of their strategy ideology, the reforms not only required functional position-orientedness to pre vail in the authority patterns of the client system, they also required the temporal orientations of the instrumen tal model. This point need not be expanded upon, for the very nature of 0 and M reforms implies the essentiality of temporal patterns in which synchronization, sequence, and 9®Gama e Silva and Pereira, op. cit.. p. 51. Translation of the author. Italics added. 203 promptness are highly valued concerns. In regard to strategy content, the change scope of the 0 and M reforms was vholistic. Indeed, as soon as the DASP was able to confine the efficiency commissions to organizational activities, a work plan was developed by the central agency, with the purpose of achieving more unity in the action of the commissions. According to this plan, all sectors were targeted: personnel (efficiency), materiel (equipment and installations), organization, work methods, financial resources, and, finally, relationships among agencies and between them and the public.99 One may find many similar symptoms of this wholism by content analyzing the successive regimentos (organizing laws) of the DASP. The reforms here analyzed adopted a prescriptive language style. The reformers took advantage of the moment and relied on a compliance model. As a matter of fact, this seems to have been an important reason for the extinction of the efficiency commissions of the military ministries, where the prescriptive approach resulted in an ^DASP, A Divislfo de Organizacao e Coordenacao e Suas Atividades em 1942 (Rio de Janeiro: Serviqo de DocumentacSo do DASP, 1943), p. 85. Translation of the author. Italics added. 204 inevitable conflict of authority. The decentralized units did not have discretion even to organize themselves and devise their work arrangements. The so-called reeimento- nadrao developed by the DASP for the efficiency commis sions established a uniform and detailed organization for all the commissions without regard for possible differ ences among the several ministries.*'®® As will be shown below, the motivational basis of such a centralization and uniformization is highly doubtful. The prescriptiveness, rather than suggestiveness, of the reforms may still be illustrated by another example. This regards the way in which any given change was to materialize in any given ministry, as a result of the work of the efficiency commissions. The usual pattern of the change process involved three phases. The first was the listing of the activities performed by the unit to be reorganized, which was done by the efficiency commis sion. The second phase was the planning of the reorgani zation to be done, which was a responsibility of the Division of Organization of the DASP. Should complemen tary data be needed, the efficiency commission would be lOO0ecree 9941, Coll.. May 27, 1942. 205 called upon to obtain them. Finally, the third phase was the presentation of a reorganization proposal to the President by the DASP. Such a proposal was usually approved and transformed into the new reeimento of the unit under analysis, by force of a decree or a decree-law, whichever was necessary.Observe that, whether the DASP*s proposal fulfilled or frustrated the aspirations of the personnel in charge of the unit being reorganized, the change was imposed upon them. This means that, in many cases, a ministerial efficiency commission might initiate action to improve the performance of a given unit and actually be penalized by having to accept changes which were undesired by that unit. Conflict and resentment were often the results of such a strategy. The examples above lead to still another charac teristic of the reforms, the cognitive orientation of their obstacle-mapping process. Valuational obstacles to change were overlooked by the reformers. This is apparent from the examination of the objectives of the Federal Civil Service Council and the former efficiency commis- ^■^Gama e Silva and Pereira, op. cit.. p. 49. sions,^2 as well as of the Division of Organization and Coordination of the DASP and, more recently, of the Ser vice of Administrative Rationalization, which is part of the present Division of Budget and O r g a n i z a t i o n .^3 The analysis of those documents, and of the messages which justified them as well, reveals that both the diagnosis of, and the remedies for, the deseases of the Brazilian public service were based upon cognitive factors. The deseases— inadequate communication networks, red tape, lack of coordination— were to be cured by the rectifica tion of workflows, by the simplification of routines, and similar medicines. The reformers apparently did not con sider that they might be taking symptoms as causes, or that the effectiveness of their prescriptions might depend a great deal upon the willingness of the patients to take them. Finally, in regard to strategy behavior, the 0 and M reforms focused action upon the formal structure. Since this has been shown in the course of this analysis, it 102See pp. 197-98. ^^Decree 20.489, Coll.. January 24, 1946, Art. 17. 207 seems unnecessary to stress this point. One might argue, however, that 0 and M reforms, by their very nature, are oriented to the formal elements of the client system. Though in practice this seems to be true, it does not need to be that way. At least in principle, even the most formal aspect of administration may be "behaviorized." The task of integrating the formal and informal aspects of administration is presently the greatest challenge of organization theory, a task which the so-called "structur alists"^^ have already undertaken. The case of work simplification (an 0 and M activity) is especially worth mentioning here. Actually, the technique of work simpli fication was devised as a strategy to integrate the formally-oriented time and motion studies with behavioral concerns such as motivation and participation. Allan Mogensen, who developed the technique, "... used the principles of motion study originally developed by the Gilbreths for structuring a program in which every member ^*The denomination of "structuralists" is, in this author's opinion, the most unfortunate description of this new school of organization theory. Some top names of this movement— which is still imprecisely defined— are Bennis, McGregor, Blake, Etzioni, Argyris, and Likert. of the organization might participate.11^^ According to Mogensen* s conception, the central agency would train field personnel in work simplification and provide them with all the assistance they needed, but actual decisions and changes were to be done independently from the central agency. This has not been the case in the Brazilian experience. Reforms have not been developmental for the field units; rather, they have inhibited initiative, risk- taking, and willingness to accept responsibility. On the other hand, they have generated much resentment— brought to the fore at the fall of the dictatorship— apathy, indifference, and resistance to change. Here is where serious doubts may be raised as to the motivational adequacy of a strategy based on centralization and con trol, as the one which has prevailed in the Brazilian experience of administrative reform. Other attempts were made by the DASP to reinitiate functional reforms in the public service, after President F. Goodwin, "Work Simplification," Factory Management and Maintenance. CXVI, No. 7 (July, 1958), 72. The expression work simplification was coined by Erwin H. Schell. 209 Vargas returned to power in 1951.106 These, however, were not as significant as the previous undertakings. The strategy remained the same, and the results as well— a wide gap between objectives and achievements. The Structural Reforms This segment of the Brazilian reform movement involves those attempts at changing the architectural design of the federal government. Though such reforms were also concerned with the functional aspects of the administration, their main focus was the structure of the federal administrative system. Another distinction between these and the functional reforms regards the IO^dasp, Exposicao de Motivos No. 670. June 11, 1951. 107For an excellent discussion of the failure of the 0 and M reforms in Brazil, see: Hugo dos Santos and James Farout, "Estudo das Atividades de 0rganiza9ao e Metodos no Service Publico Federal" (unpublished paper of the Organization and Methods Service of DASP, Rio de Janeiro, March, 1960). (Mimeographed.) This was an empirical study involving eight ministries. It becomes clear, from the authors* observations, that the 0 and M reforms were patterned after the instrumental model of change strategy used in this study. See, as a complemen tary source: Fabio de Carvalho Alves, As Atividades de Organizacao na Administrac^o Pufelica Brasileira (Rio de Janeiro: Imprensa Nacional, 1950). In the English liter ature, see: Siegel, "The Vicissitudes . . . ," pp. 205- 14. 210 hierarchical levels at which they were directed. The structural reforms have been concentrated at the highest echelons of the system, the Presidency of the Republic and the ministries. The functional reforms have revolved around middle and lower management levels. Hence the con venience of distinguishing between these two types of reforms for analytical purposes. Though it shares with all the others the strategic characteristics of the instrumental model, the structural reforms are unique as to one point at least— they have never won a political breakthrough. Indeed, initiated by President Vargas in 1952, the structural reforms— usually termed "administrative reforms" in Brazilian circles— are the best example of "the vicissitudes of governmental reform in Brazil."*®® Thus, they lend themselves to a study of strategies of political breakthrough, more than to a study, like the present one, of strategies of imple mentation of administrative changes. The distinction is relevant, for many variables central to the first kind of study (pressure groups, political majority, et cetera) may *°®Siegel, "The Vicissitudes . . . 211 be of a secondary importance to the latter. The former is principally a study of planned organizational change, while the latter deals more with political change. However, as said before, it may well be that some of the very characteristics of the instrumental model which have encumbered the implementation of the reforms have also accentuated their political difficulties. For example, the globality of the reforms imposed administra tive demands which the bureaucracy was unable to fulfill; from another perspective, that characteristic was also responsible for many political barriers against the pro posed changes. Resistances were elicited which might be dormant under a gradualistic strategy. Therefore, it seems worthwhile to provide a gener al description of the successive attempts at changing the structure of the federal administrative system, so as to identify the strategic characteristics of such reforms. Historically, the structural reforms were begun in October of 1952 when President Vargas, in a public speech, observed that "... the multiple and complex problems of national interest demand a profound reform of the adminis- 212 trative system of the Union.M*09 Shortly afterwards, the President charged a group of specialists, almost all from the DASP, with the responsibility of preparing the reform project which was finished before the end of 1952. A great connoiseur of the Brazilian political process, Vargas devised a strategy to enhance the possi bilities of legislative approval of the reform. Thus, he personally appealed to all parties to support the govern mental project. As a result of his interest and action, eight political parties formed, on an extra-official basis, the so-called Inter-Party Commission for the Federal Administrative Reform. Each party had members of both the higher and the lower houses of Congress in the Commission. And on December 20, 1952, the Commission received from the hands of the President the reform project. In terms of political strategy, this was a rare example in the Brazilian experience. It is even more 109 As quoted in Comiss^o de Estudos e Projetos Administrativos, op. cit.. p. 35. ■^^Comissao Interpartidaria de Reforma Adrainis- trativa Federal, Reforma Geral do Sistema Administrativo da Uniao (Rio de Janeiro: Departamento de Imprensa Nacional, 1953), pp. 3-4. 213 significant because it reveals a high political adapt ability on the part of Vargas. The man who signed decree- laws was showing himself able to adjust to, and manipulate, the tactics of a new political ideology. One might wonder, however, whether the reformers could do the same. The Inter-Party Commission held fourteen sessions for the examination of the governmental project and ended its activities in May, 1953. The final report of the Com mission was sent to the President with the pertinent suggestions.*** The majority of those suggestions was incorporated in the original project. For the reformers, the President "... made too many concessions. "H2 Nevertheless, the project was sent to the House of Repre sentatives on August 31, 1953.*^ It is reasonable to expect that after such tact fulness , the reform proposed by Vargas would have a smoother legislative path than is usual in cases of m Ibid.. pp. 3-51. H2cePA. o p . cit.. p. 53. Translation of the author. 113 Brazil, House of Representatives, Bill No. 3563 (1953). 214 reform proposals. After all, possible sharp differences had been cleared before the final version of the bill by the Executive. The legislative process should be more expeditious. However, this was not the case. Vargas died one year later. The project outlived many legislative commissions and five Presidents of the Republic. Indeed, eight years later, in March of 1961, even the first dis cussion of the bill had not been completed. From the standpoint of the strategic implications of its content, the reform proposed in 1953 corroborates what has been said in regard to the other reforms. The elements of the instrumental model of change strategy may be easily identified in the document. For the sake of avoiding repetition on this matter, and also considering that the foregoing analyses of the other reforms may allow, at this point, a less detailed discussion of the strategic model of the reforms now under examination, a few comments may suffice to characterize the instrumental ism of the 1953 reform. First', it left untouched any single agency of the federal government (wholism). H4For a detailed account of the legislative pro cessing of this project, see: CEPA, op. cit., Chapter IV. 215 Second, the two basic principles which guided the elabora tion of the project were (1) homogeneity of functions and (2) span of control,which is indicative of the philosophy behind the reform. It is from the justifica tion of the project, however, that one may infer the model of change strategy which would have been applied to imple ment the reform, had it been approved. Indeed, the following excerpt leaves no doubt as to the instrumental ism of the reform. After reiterating the need for over hauling the administrative machinery in face of the increasing demands of the country, the reformers observed: There emerges, then the inescapable question: what to reform? To this fundamental question there are tens of possible answers, easily* All of them may be sub- summed into only three: to reform the structure, to reform the functioning, to reform the men. The last implies a revolution in the psychology of the public servant, in his attitude towards the public, a revolu tion of a spiritual sort to be brought about gradually and that is clearly excluded from the scope of this prolect.t!6 To this author, it does not seem necessary to insist on the analysis of the 1953 reform. Its focus of ^-*Ibid.. pp. 42-43. ^-^Comissao interpartidaria de Reforma Adminis trativa Federal, op. cit.. p. 59. Translation of the author. Italics added. 216 action was the formal structure only; its obstacle-mapping process was geared to cognitive limitations alone. One by i ■ one, the characteristics of the instrumental model of change strategy may be identified in it. The second structural reform dates from August, 1956. That year President Juscelino Kubitschek created a Commission for Administrative Studies and Projects better known as CEPA. The membership of this new group also came from the DASP. CEPA worked under all sorts of difficulties. Its members were understaffed, ill-housed, and not even paid for their services to the Commission. Nevertheless, CEPA held a total of seventy meetings between its installation in August, 1956, and its adjournment in December, 1960. A good number of projects which were submitted to the President resulted from these m e e t i n g s .U S Actually, the work of CEPA was a continuation of the former project presented by President Vargas to Con gress in 1953. There were certain refinements, but the essence was the same. Thus, among the projects prepared l^Decree 39855, Coll.. August 24, 1956. II^Cepa, op. cit., Chapter IV. 217 by CEPA were the streamlining of the Presidency of the Republic,the separation of the Ministry of Justice and Internal Affairs into two, a Ministry of Justice and a Ministry for Internal Affairs, and the creation of new ministries, most of which had been included in the 1953 reform. The final recommendations of CEPA were not signi ficantly different from those of its predecessor. For example, the recommendations relating to the structure and functioning of the Presidency of the Republic revolved around three points, namely, the reduction in Presidential contacts, the remodeling of the Presidency staff, and the reinvigoration of the DASP.^20 it is unnecessary, there fore, to dwell on the analysis of the strategic implica tions of CEPA's projects. Nevertheless, it is appropriate to observe that two of the Commission's proposals endorsed by President Rubitschek were finally approved by Congress. These regarded the creation of the Ministry of Mines and 119 This project comprises the first volume of the CEPA's publications. See: CEPA, A Reforma Administrativa Brasileira. Vol. I: Reorganizacao da Presid^ncia da Republica (Rio de Janeiro, Departamento de Imprensa Nacional, 1960). ^^Ibid., pp. 145-56. For the complete recom mendations, see ibid.. Chapter V. 218 Power and the division of the former Ministry of Labor, Industry and Commerce into two distinct entities, the Ministry of Labor and Social Welfare and the Ministry of Industry and Commerce. These were the only efforts of the CEPA that had any practical consequences. With the succession of Kubitschek by Janio Quadros, the structural reforms came to a standstill. When the latter resigned seven months after he took office, a parliamentary regime was instituted in Brazil. A new effort to reform the structure of the federal admin istrative system was then sponsored by Prime Minister Brochado da Rocha. It was an abortive attempt, among other reasons because President Goulart's only thought at the time was to promote the return of the presidential regime. The most recent structural reform encompassed by the present study dates from 1963 when President Goulart, now with full presidential powers, created a new reform group to be headed by an Extraordinary Ministry for Administrative Reform. As happened with the Hoover 121i)ecree 51705, Coll.. February 14, 1963. 219 Commissions in the United States, the reform took the name of its main conductor, Amaral Peixoto. The difficulties faced by the government in 1963- 64, which culminated in the overthrow of President Goulart by a civilian-military c o u p d'etat in April of 1964, left no hopes for the Amaral Peixoto Reforms. In spite of the fact that a special ministry (though without portfolio) was created for it, the reform never had any of its pro jects approved. The Amaral Peixoto Reform was again a continuation of the works first proposed in 1953 by President Vargas and later refined by CEPA. Indeed, in a speech before the House of Representatives, Minister Amaral Peixoto recog nized: "... the effort of administrative reform that we are developing now is a continuation of CEPA’s works."*22 This had to be so, for the group in charge of the reform was Dasoian in origin, and its philosophy of change and diagnostic orientations were essentially the same as the previous undertakings. This time, however, ^22Mj_n- j _ stro Extraordinario para a Reforma Adminis trative, "A Reforma Administrativa Federal de 1963," A Speech before the House of Representatives, April 17, 1953, by the Extraordinary Ministry for Administrative Reform, p. 19. (Mimeographed.) 220 the group was much larger (150 people), better staffed and housed, and with considerable financial resources in com parison with its former counterparts. Thus, in the fashion of the United States Hoover Commissions, a number (twenty-three) of task forces were formed, one for each ministry and several others for specific problems such as salary policy, the reinvigoration of the merit system, and the revision of work procedures and methods in the federal service. Insofar as a behavioral orientation to change is concerned, the only point where the Amaral Peixoto Reform differs from the previous works is in the import ance attributed to training as a condition of success for the reform. Specifically, the report of the task force on the strengthening of the merit system reveals a clear understanding of the strategic failures of the reform movement in Brazil. Though the recommendations of that group were never implemented, it is worthwhile to quote its remarks in regard to training: The attempts to increase efficiency in the Brazil ian public administration have centered almost always on the organizational aspect. There has been insist- 193 * For the composition of all such groups, as well as their general policies, see: Ministro Extraordinario para a Reforma Administrativa, Plano de Execucao da Reforma (Rio de Janeiro: Grafica Editdra Livro, 1963). 221 ence on remodeling structures and distributing legal competences, integrating and subdividing agencies, autarchies, services, et cetera, but very little has been done, in fact, for the professionalization of the public servant. Hence the stoppage of the attempted modifications by the indifference or even resistance of the employee, so that such modifications do not produce results which are proportional to the effort and resources applied to them. The fundamental cause of the continuous failures is the human element. The effort developed for the sake of administrative efficiency is wasted by the dilluting action of the functionary himself, usually resulting, if not in a negative balance, at least in the conservation of the status quo. Therefore, it is urgent to expand and intensify the efforts toward increasing the professionalization of public personnel, that is, toward training and development at all levels. In order for the public service to comply effectively with its social mission, it is necessary to promote the formation of a progres sive mentality, based on professional competence, leadership development, organizational loyalty, and sense of team, to mention only some of the most important elements. Following such a diagnosis, the task force on personnel presented a full program of training to be developed concomitantly with the implementation of the 125 reforms so as to facilitate such implementation. * As ^Ministro Extraordinario para a Reforma Admin istrativa, "Normas para Preservasao . . . pp. 16-17. Translation of the author. Italics added. 125Ibid.. pp. 17-29 said before, this is, however, the only significant devia tion of the Amaral Peixoto Reforms from their predecessors, as far as a behavioral orientation is concerned. Also, if compared to the whole product of the reform group, this part corresponds to less than 5 per cent of the total. The simplification of the Presidency of the Republic, for example, was again approached from the same perspective of reducing the existing formal relationships between the President and administrative agencies, as if such a measure would actually result in less direct contacts for the President. The behavior would be the same, so that, informally, the load upon the President would continue unaltered. To use the reformers* jargon, the span of control would remain violated. The same sort of assump tion may be inferred from statements like the following one, which refers to one of the guidelines for the work of the task forces: "Not a single act practiced by public servants is to result from improvisation, but from pre vious decisions, matured and articulated into plans."126 ^^Ministro Extraordinario para a Reforma Adminis trative, Reforma Administratlva_Federal de 1963 . . . , p. 19. Translation of the author. 223 From a behavioral standpoint, such a prescription is pointless. From a strategic one, it implies a disregard for the ecology in which the reform is to operate. The present government of Castello Branco has al ready engaged in the task of reform. A new group was formed, now without connections with the DASP, as in the former efforts. Under the direction of Helio BeltrSo, a well known expert in business administration, this new group seems to be taking a different strategy than the one that has prevailed in the Brazilian experience. Though this new attempt— still under way— is outside the scope of the present study, it is worthwhile transcribing a public remark by the major person responsible for the reform. After justifying his rejection of the projects prepared by the Amaral Peixoto staff, Beltrao observed: It is useless simply to legislate, alter the ministries, and distribute their functions. This is a problem of organizational technique, one of the aspects of the matter. What we need is a change of mentality, as one among the several factors to be utilized in the r e f o r m . 127 "Beltrao na URGS— Re forma administrativa fed eral nao inclui o Ministerio da Previd£ncia,n Correlo do Povo. September 25, 1965. Translation of the author. Italics added. 224 This statement is clearly indicative of a change in the strategy of administrative reform. As in the case of the task force on personnel of the Amaral Peixoto Reform, it reveals a concern, until then nonexistent, with the administrative ecology of the reform, with the behav ioral orientations of the client system, and with the need for harmonizing a strategy of change with the behavioral profile of the bureaucracy. To the extent that it is actually so, new horizons— more promising ones, we venture to anticipate— may be open to the enterprise of adminis trative reform in Brazil. Conclusion In our effort to demonstrate the failure of the Brazilian administrative reforms, we may have produced too pessimistic an analysis. Such was not our intent at all. If there was actually an overemphasis on the negative aspects, this must be taken as one more of the frequent happenstances in which the social researcher overlooks data that might reduce the strength of his con clusions. In other words, the intention was not to convey 225 the impression that there was a complete failure of the reforms. It would be both senseless and highly unfair to suggest that the DASF did not contribute to the improve ment of the Brazilian administration. Indeed, its contri bution was a very important one. The contention that underlies this study, however, is that the reforms fell far from their proposed objectives, and this— the author hopes— has been demonstrated in this chapter. A second contention, less explicit than the first, is that more could have been done if a different change strategy had been adopted. The fact that such a point can be neither rejected nor sustained, for the past is gone, may raise against it the blemish of pure academicism. It is not so, at least for two reasons. First, Brazil is again under a strong regime, one where the compliance model is viable as in Vargas' time, To a certain extent, this represents a historical return, insofar as some political conditions of the past have been reinstated. Therefore, an attempt to contrast the fugacity of the achievements of a compli ance approach with the endurance of the attainments of an internalization one, and to demonstrate the inviability of an instrumental model of change strategy may be of extreme 226 relevance if it influences today's " r e f o r m m o n g e r s . " 1 ^ Second, Brazil is likely to be subject to new strong regimes in the near future. Hence the importance of criticisms to the strategy of change used in previous but similar conditions. In short, this rejection of the instrumental model may be a warning against the adoption of a strategy whose possible fruits are bound to be lost as soon as the prevailing political ideology is replaced, as occurred after the fall of the dictatorship in 1945. Though the mistake was done once, it need not be repeated. Another objective of the present chapter was to demonstrate that, as a whole, the reforms did not apply different strategies. Rather, directed by the same group;, the reforms were patterned after one prevailing strategy. Such a strategy— and this was the main purpose of the chapter— conforms to the model of change here called instrumental. Its characteristics, as to strategy con tent, are a wholistic change scope, a prescriptive language style, and an obstacle-mapping process geared to cognitive, rather than valuational, obstacles. As to 128jjor(j coined by Hirschman, op. cit., p. 276, 227 strategy behavior, the focus of action was on formal elements; and as to strategy ideology, given patterns of authority relationships and temporal orientations were , 4 , required from the client system. f CHAPTER IV AUTHORITY PATTERNS AND THEIR IMPLICATIONS FOR CHANGE STRATEGY: SOME THEORETICAL PROPOSITIONS The preceding chapter analyzed the experience of administrative reform In Brazil, from the standpoint of the strategy of change which has guided that experience. In- order to assess the adequacy of such a strategy, it becomes necessary to understand the behavioral character istics of the client system where the strategy has been applied. The main purpose of the present chapter is to relate patterns of authority to the induction of change, as well as to lay down the theoretical foundations and make explicit the assumptions which underlie the rationale of the study of authority patterns carried out in the next chapter. The content of the present chapter is divided into six sections. The first focuses on the analytical levels 228 at which authority has been studied, and criticizes the predominance of a normative, rather than a behavioral, approach to the phenomenon. Hie second section develops some theoretical propositions as to the implications of authority patterns for the devising of strategies of change. The following section deals with the specific dimension of authority behavior with which this study is concerned, namely, position-orientedness; it also relates this dissertation to previous works which dealt with that same dimension. The fourth section presents and discusses the indicators to be used in the analysis of position- orientedness in authority patterns, which constitutes the main purpose of Chapter V. Since position-orientedness is neither "good” nor "bad" in itself, it is indispensable to distinguish between position-oriented behaviors which are functional to, or supportive of, the ideology of the instrumental model of change strategy, and those which are dysfunctional to, or disruptive of, that ideology. The issue of functionality and dysfunctionality of position- orientedness in organizational behavior is examined in the fifth section of this chapter. Finally, since this study adopts a comparative approach, an attempt is made in the 230 closing section to relate this work to the current trends in comparative public administration. Authority Relationships: The feed_for Behavioral Approaches It is a common— and intellectually sound— rule of academic tradition to devote a section on "the importance of the study" to any serious endeavor of scholarship* When the subject is authority, however* the rule is entirely dispensable. In the words of Humayun Khan: A study which purports to focus on a phenomenon like authority, however, scarcely seems to require justification on grounds of importance, for its importance is self-evident. What it does require is justification on grounds of originality, for author ity, in one sense or another, has been the concern of all thinking men from the beginning of time, and is, as one adviser told the writer, a "heavily overworked field."! Indeed, because of its pervasiveness, authority has been a recurrent theme, throughout the times, to writers from all walks of knowledge. From romance*’ to ^Mohammed Humayun Khan, "Authority Behavior: An Empirical Approach to Comparative Public Administration" (unpublished dissertation, University of Southern Calif ornia, 1965), p. 5. 2victor Hugo's Lee Miserables. trans. Charles E. Wilbour (Hew York: The Book League of America, 1943), is 231 educational psychology, from studies on juvenile delin quency to political science, the literature on authority is simply copious. For the purposes of this dissertation, however, more important than the scarcity or abundance of literature are the analytical levels in which the subject has been studied. More precisely, how much of grand theory,^ as contrasted with operationality, does the literature on authority present? How measurable is the concept today? According to Bierstedt,^ the concept of authority has been studied at three levels: (1) the philosophic, (2) the "mundane-political," and (3) the sociological. The first two deal with the subject on a highly intangible an excellent example of an intra-role conflict imposed by a position of authority. Inspector Javert finally found in his suicide the only escape from that conflict. For an excellent and rather comprehensive review of the literature on authority, see Khan. o p . cit., Chapter III. ^See footnote 18 of Chapter II. K. Bierstedt, "The Problem of Authority," Freedom and Control in Modem Society, ed. Morroe Berger, Theodore Abel, and C. H. Page (New York: Van Nostrand, 1954), pp. 67-81. 232 dimension, such as Che divine authority of God, or the "authority of the law"; the sociological level deals with organizational authority, and, as pointed out by Bierstedt, is the least researched of all. Notwithstanding the fact that the sociological is far more operational than both the philosophic and the "mundane-political" levels, it becomes clear from any cursory survey of the literature that, even at that level, the concept of authority has been approached as subjec tively as in Mary Follet's "authority of the situation,"** or as in the "authority of ideas" and the "command author ity" attributed to staff and lipe people, respectively.^ Those themes were typical of the administrative literature in the thirties and forties, and still inspire many works by the so-called "management scientists" today. Unfortunately, the great majority of works on authority— even the most recent ones— has failed to bring ^Mary Parker Follet, Dynamic Administration: The Collected Works of Marv Parker Follet. ed. H. C. Metcalf and L. Urwick (New York: Harper & Bros., 1940), Chapter II, particularly pp. 58-64. ^See John Pfiffner and Frank P. Sherwood, Adminis trative Organization (Englewood Cliffs, N.J.: Prentice- Hall, Inc., 1960), p. 178. 233 the concept to an empirical ground so as to permit research and comparability on authority relationships, A remark that can be generalized is made by Dwight Waldo, in a review of Authority, a book edited by Carl J. Friedrich: Although the tone is descriptive and analytical, not philosophical or ideological, the propositions set forth are for the most part too general to serve as the basis for experimental research, or beyond rigor ous investigation because of passage of time or absence of data. . . .8 After a comprehensive survey of the literature on authority for the purpose of writing his doctoral disser tation, Humayun Khan concluded that: . . . In those areas of public administration where the concept has been used for purposes of analysis and research, enough attention has not been paid to developing the more pertinent and the more patent aspects of authority. Exploration of the phenomenon has stopped at the level of qualitative analysis, which examines issues like "the sources of authority" and "the nature of authority," or mechani cal analysis, which concerns itself with the "struc ture of authority" and its distribution in various organizations. There are few attempts at discovering how authority manifests itself in organizational relationships. In other words, the crucial question, how do persons in positions of authority in an D Dwight Waldo, Review of the book Authority. ed. Carl J. Friedrich (Cambridge: Harvard University Press, 1958), in Administrative Science Quarterly. Ill, No. 2 (September, 1958), 268, 234 organization behave when they exercise that authority, has seldom been asked.9 The study carried out in the next chapter approaches the concept of authority from a behavioral per spective, so as to make possible the measurement of the phenomenon as it occurs. It is believed that attempts such as that can increase substantially the usefulness of the concept, by lending to it a much greater analytical power as a research instrument, and also a much greater explanatory power as a theoretical construct. Authority Patterns and Planned Organiza tional Change The major theoretical proposition guiding the empirical analyses which are carried out in the next two chapters is that the greater the compatibility between the requirements of a strategy of change as to the behavioral patterns of the client system, on the one hand, and the actual prevailing behavioral patterns of that system, on the other, the greater the probabilities of success of the intended change. Conversely, the lesser that compatibil- ^Khan, o p . cit.. p. 3. 235 ity, the lesser the probabilities of success. Statistically, the probabilities of success of any intended change vary directly with the degree of compat ibility between the strategy ideology of the change model and the behavioral profile of the client s y s t e m , ^ Graphically, this proposition may be represented as in Figure 5. Since it stands at a very high level of abstrac tion, such a proposition is applicable to all dimensions of organizational behavior, one of the most conspicuous of which is the dynamics of authority. "Indeed, the organization may be defined as a relatively permanent system of authority relationships."^ As a matter of fact, these relationships are not confined to formal organiza tions at all; they also exist within informal groups like lOobviously, the fate of any change attempt depends on many more variables (political support, for instance) than the one examined in this chapter. Further more, even the behavioral profile of a client system is composed of many traits besides authority relationships. For the sake of manageability, however, only this trait and temporal behavior— incidentally, two crucial aspects of the whole profile— are considered in this dissertation. ^Robert Presthus, The Organizational Society (New York: Alfred A. Knopf, 1962), p. 140. 236 ./v PQ T> e < u ■ u 0) X * >s AJ X > •H 4J < d I o c •H 4J CO 0) £ 1 o y„.- Success probabilities of intended change s/ o 0 f t ) 4J f t ) CO £ f t ) AJ 4 - 1 f t ) e f t , £ ) 0) U O <8 J. v O ') Fig. 5.— Success probabilities of intended change, in terms of the strategy ideology of the change model and the behavioral profile of the client system. 237 the family, between sons and parents, and, in many cul tures, between the spouses as well. In short, authority is an almost omnipresent phenomenon in man’s life. There fore, a proper understanding of the behaviors and values that pattern what Caplow calls "the interaction between unequals"12 is indispensable for the understanding of the administrative process itself. When one turns to planned organizational change, authority relationships become a crucial strategic factor, for any large-scale change attempt depends considerably upon those who hold the chain of command in the client system. In the case of governmental reforms, the transla tion of proposals into accomplishments depends more on the executives than on any other single group of the civil service. Several factors account for the critical import ance of the executive group. However, suffice it to say that it is through the hierarchical grid that the change effort is usually channeled. From the top to the bottom, the executive^ is the main linkage between the change agent and the client system. Besides, more than any other, the •^Theodore Caplow, Principles of Organization (New York: Karcourt, Brace & World, Inc., 1964), p. 110, 238 position of the executive ascribes to its incumbent a double role in the change process— he is concomitantly an agent and a client of the intended change. He holds the responsibility for effecting change while being affected by it. The crucial importance of the executive group for governmental administrative reforms suggests a whole series of key questions about strategy of change. For instance: Which expectations or values about executive behavior tend to prevail in the client system? How can such values be manipulated in order to facilitate the change process? Is it possible for a particular strategy to overlook the predominant values about executive behav ior in the client system? If so, how consequential may such a flaw be for the fate of the change attempt? What are the prevailing patterns of executive behavior in the client system? Are there strategies which depend on given behavioral patterns which do not prevail at all among executives in the client system? If so, how can this affect the outcome of the intended change? It is around those questions that the rationale of the study of authority patterns to be carried out in the next chapter is developed. Thus, the main concern is with the adequacy of a given change strategy in view of the prevailing patterns of executive behavior and of expecta tions about such behavior, in the client system. On the basis of the general proposition stated at the beginning of this section, it is submitted that the adequacy of a change strategy varies directly with the degree of compat ibility between the requirements of that strategy as to the prevailing patterns of executive behavior and of expectations about such behavior, in the client system, and the patterns that actually prevail in that system. Again, the greater the compatibility, the more adequate the strategy, and, consequently, the greater the likeli hood of success of the change attempt. It may appear that this study overestimates the role of executives in the change process. It can be argued, for instance, that any member of the client system, and even the change agent himself, is always both subject and object of change. The client system is not entirely passive; it is not a tabula rasa at all. Nor is the change agent impermeable; rather, he is also influenced by the client system. After all, change, as learning, is not 240 a unilateral but a reciprocal process. This is one of the most warranted conclusions that can be drawn from systems theory and from psychological theories on human inter action as well. Nevertheless, from the perspective of role theory,^3 the executive differs rather sharply from both the change agent and the subordinates in the client sys tem. The distinction regards the extent of role conflict that each of those positions imposes-on its incumbent.. For subordinates are not likely to experience any serious psychological stress by lining up with their work group against the intended change, whenever that change is rejected by the group; their loyalty orientations are relatively well defined and stable. The same can be said of the change agent; he is loyal and committed to the ■^Both the terminology and the definitional con tent of role theory used in this chapter come from: Neal Gross, W. S. Mason, and A. W. McEachem, Explorations in Role Analysis (New York: John Wiley & Sons, 1956). For the definition of concepts such as role, position, incumbent, and expectations, see their Chapter IV, parti cularly, p. 67. For the conceptualisation of role con flict and related notions, see their Chapter XV, particularly, pp. 248-49. ) 241 reform.^ The executive, however, faces all the dilemmas of the "man at the m i d d l e . " ^ There is not a clearly preferable choice for him. He owes loyalty to both his group and his superiors. At least theoretically, none of his roles— as an agent and as a client of the change process— visibly predominates over the other. He plays both with equal intensity and commitment. Therefore, his l^There are exceptions to this pattern, and they may not be too rare. However, this seems to be the pre vailing orientation, as shown by studies on organizational group behavior. Among the inumerous pieces which present evidence on group loyalty, the reader may consult the following selected ones: Elton Mayo, The Human Problems of an Industrial Civilization (New York: Viking, 1931); F. J. Roethlisberger and W. J. Dickson, Management and the Worker (Cambridge, Mass.: Harvard University Press, 1939); S. A. Stouffer et al.. The American Soldier (Princeton, N.J.: Princeton University Press, 1949), Vols. I and II; Chris Argyris, Personality and Organiza tion (New York: Harper & Bros., 1957); and Rensis Likert, New Patterns of Management (New York: McGraw-Hill Book Co., Inc., 1961). The first two books refer to the famous experiences at the Hawthorne plant of the Western Electric. The third one deals with group loyalty in the armed forces. The last two books are good sources of reference, for they review and summarize the empirical literature on the matter. For an insightful comment on group-loyalty behavior in the Brazilian bureaucracy, see: Frank P. Sherwood, "The View from the Outside," Public Administra tion Review. V, No. 4 (December, 1963), 247-52, particu larly, p. 251. l*5]?remont A. Shull and Delbert C. Miller, "Help the Man in the Middle," Nation1s Business. LI, No. 2 (February, 1963), 36-37. 242 role conflict increases in direct relation to the degree of opposition that the intended change may raise in the client system. For all these reasons, plus the fact that it is the executives who are entrusted with the authority and charged with the responsibility for implementing the reforms, an empirical study of planned change such as the present one cannot neglect the analysis of how those in positions of command actually behave, lest much of the explanatory power of the study is to be lost. After all, "reactions to authority constitute the most critical variable in organizational accommodation."^ Now, as advanced before, in order to analyze how authority is exercised, it becomes relevant to identify the prevailing expectations about the behavior of execu tives. The relevance of this point derives from the fact that one's behavior does not occur in a vacuum but in a social space which may vary from situation to situation, and from culture to culture. To soma extent, one's behav ior is situationally conditioned. Therefore, expectations come to be an important variable for the explanation of the behaviors of those in command. Along this line of 16Presthus, op. cit., p. 140. 243 reasoning, the next chapter deals with how executives be have, and how they are expected to behave, in the Brazil ian bureaucracy. Position-Orientedness in Authority Re lat ionships The analysis of authority patterns in the Brazil ian bureaucracy focuses on a distinguishable feature of the phenomenon of authority, namely, position-orientedness. As implied in its name, the concept of position-oriented ness revolves around the question: To what degree is the ’’interaction between unequals” governed by factors such as occupancy of a position, status, hierarchy, as against competence and interpersonal relations? This is a criti cal question for the objective of the next chapter, i.e., to probe into the correctness of the assumptions of the instrumental model of change about the prevailing patterns of authority relationships in the Brazilian bureaucracy. Since those assumptions center on position-orientedness, this aspect of authority relationships had to constitute a basic point of the study. 244 As far as position-orientedness is concerned, a behavioral approach to authority should consider questions such as the following: (1) Are there behaviors which rely predominantly on positional factors for their acceptance? If so, which are they? (2) How does the socialization process relate to position-orientedness in authority relationships? (3) How does position-orientedness relate to organizational health? (4) More precisely, is position-orientedness dysfunctional by definition, or may it affect positively the administrative dynamics? (5) What consequences derive from functional and dysfunctional position-orientedness for the organization's ability to change? (6) Are there types of organizations, or cultural settings, which tend to be particularly resistant, or receptive, to position-orientedness? If so, what are the implications of the fact for strategies of change? This research is concerned with the patterns of behavior that characterize authority relationships in Brazil, and with the implications of those patterns for the administrative reforms attempted in that country. It can be viewed as an application of earlier studies on positional authority to strategies of planned administra- 245 tive change. Those studies were first developed at the Youth Studies Center, of the University of Southern California, in the field of juvenile delinquency. On the other hand, as an outgrowth of his observations on administrative behavior in Pakistan, Professor Frank P. Sherwood developed a preliminary research on authority orientations in Brazil.^8 Later, Mohammed Humayun Khan, working under the guidance of Professors Sherwood and McEachem, expanded those ideas and applied them in a comparative study of authority behavior in Pakistan and the United States.*9 Finally, this research attempts to apply Khan's contributions in a completely distinct theoretical framework, and in a somewhat different methodological approach. *?See Alexander W. McEachem and Ted Palmer, "Effectiveness of Different Treatment Processes," in Views of Authority: Probationers and Probation Officers, ed. A. W. McEachem (Los Angeles: Youth Studies Center, 1961), pp. 190-212. *®Frank P. Sherwood, "Comparative Authority Orien tations in the United States and Brazil: A Preliminary Probe" (unpublished paper delivered at a Workshop Confer ence for Trainers on New Research Findings in the Behav ioral Sciences held at the University of Southern Calif ornia, Civil Center Campus, Los Angeles, April 30, 1964). 19Khan, op. cit. Professors Sherwood and McEachern served in Khan's Doctoral Dissertation Committee. 246 Though relying on the same questionnaire, and, 20 in regard to the United States, the same data, this dissertation bears little direct relationship with the one by Khan. First, as to focus of analysis, Khan was primarily concerned with probing into the possibilities of a behav ioral approach to authority in comparative public admin istration; and, secondly, his intent was to use such a behavioral approach to authority as a means of testing Riggs' notion of "overlapping" as well as the logical con sistency of one of the characteristics of Weber's ideal- t y p e . 22 t h e other hand, the focus of analysis of this dissertation is on strategy of planned administrative change, so that authority relationships are relevant only 2^This study uses Khan's questionnaire on author ity behavior. The author translated that instrument from English into Portuguese, so that it could be applied in Brazil. Data on Brazil were collected in 1964 but never used, in any way, by Khan or anyone else. 2-^See his major hypotheses on page 15, and minor hypotheses on page 16. Khan, op. cit. 22gee his minor hypotheses on pages 17 and 18, as well as the discussion on Weber's model on pages 264-68. Khan, op. cit. 247 to the extent that they help identify the adequacy or inadequacy of a particular strategy of change. Here a behavioral approach to authority is developed not for its own sake, but for helping a diagnosis of the change pro cess. In fact, authority relationships represent only one of several variables on which such a diagnosis can be b a s e d . Obviously, to these differences in purposes and analytical foci correspond not less important differences in the theoretical background of the two dissertations. From a methodological standpoint, this research design also differs from the one by Khan in that we dis tinguish between functionality and dysfunctionality of position-oriented b e h a v i o r s ,24 while he expressly ignored such a distinction. This innovation in the research design may result not only in a methodological contribu tion, but in a theoretical one as well, to the extent that a higher level of integration of the substantive content 23xhis is indicated, for instance, by the fact that the whole subject of Khan's dissertation regards only one out of the seven chapters of this dissertation. 24ihi3 is the object of a forthcoming section of the present chapter, "Functionality and Dysfunctionality of Position-Orientedness in Organizational Behavior." 248 of this behavioral approach to authority may be achieved. Authority Relationships: Indicators of the Pos ition-Orientednes s Whenever position "A" entitles its occupant to require certain behaviors from the occupant of another position, "B,” the interaction between those occupants within the social system where their positions are inter connected may be called an authority relationship. Ex amples of such relationships are those between parents and sons, teacher and pupil, probation officer and pro bationer, and superior and s u b o r d i n a t e . 2 5 From this perspective, three characteristics are essential to an authority relationship. (1) It rests on 25xhis study focuses on relationships between superiors and subordinates only. The social systems dealt with here are formal organizations, defined as social units in which ". . . the goals to be achieved, the rules the members of the organization are expected to follow, and the status structure that defines the relations between them (the organizational chart) have not spontaneously emerged in the course of social interaction but have been consciously designed a priori to anticipate and guide interaction and activities.” Peter M. Glau and W. Richard Scott, Formal Organizations (San Francisco: Chandler Pub lishing Co., 1962), p. 5. For Blau, this formal explicit ness of purpose is what distinguishes formal organizations from social organizations in general. 249 a formal, rather than informal, structure; it derives from a position, rather them from a role.26 (2) It involves an interaction in which the actors are unevenly related to each other, from a positional point of view; that is, they are formally placed in positions of super-ordination and subordination.(3) is primarily2® a compliance- evoking-responding interaction, where the superior attempts to obtain certain behaviors from the subordinate, and the subordinate reacts along a conform!ty-deviation continuum in face of the superior's demands. 2®This Is a relevant distinction, among other rea sons because it eliminates the possibility of confusion with other concepts such as leadership, power, or personal (in contrast to organizational) authority. For a clear differentiation between these concepts, see: Ffiffner and Sherwood, op. cit.. pp. 74-80; and Harold J. Leavitt, Managerial Psychology (2d ed.; Chicago and London: The University of Chicago Press, 1964), pp. 165-80. ^^Those placed in "super-ordinate" positions will hereafter be referred to as "superiors," "executives," or simply A's; those in subordinate positions will be called "subordinates," or B's. 28We say "primarily" because, in many instances, a superior's behavior is not compliance-evoking at all. Nevertheless, such noncompliance-evoking behaviors can also be position-oriented or nonposition-oriented. Since the distinction is critical for the study, noncompliance- evoking behaviors are taken as "authority behaviors" as well. 250 For the purposes of the coming study, position- orientedness refers to a pattern of authority relation ships whose referents are predominantly formal elements, such as hierarchy, position, and legal authority. For the identification of that pattern, three indicators were developed, namely: (1) status-protectiveness, (2) emphasis on hierarchy, and (3) extension of the scope of the hierarchical relationship.^ Status-protectiveness occurs whenever A, feeling that B is positing a threat to his position, resorts to acts designed to defend that position. A cotraon instance of such situations is a subordinate questioning his superior's orders, for whatever reasons it may be. In this case, A might react in at least the following alter native ways: (1) trying to get rid of B; (2) reinforcing his personal control over B; (3) judging the objection on its merits; and (4) considering such an independence as beneficial to the organization, and, therefore, giving his stimulus to B. It is clearly inferrable that alternative ^Khan called these indicators "protective behav ior," "behavior emphasizing hierarchy," and "behavior extending the scope of the hierarchical relationship (op. cit.. pp. 133-35). 251 1 indicates more position-orientedness, whereas alterna tive 4 indicates less position-orientedness. Several kinds of threats to A's position are suggested in the questionnaire, in a number of different hypothetical situations.30 Emphasis on hierarchy refers to behaviors which overlook factors like competence, skill, human relations, group morale, et cetera, for the sake of the position it self. For example, two of A’s subordinates have a quarrel with each other, and one of them, C, lower in rank than the other, B, complains to A against B. In this case, A would have at least the following alternatives: (1) He could penalize C for by-passing the "proper channels." (2) He could refuse to examine the complaint and instruct C to bring it to B’s notice first. (3) He could receive the complaint and refer it to B for his comments. And (4), he could receive the complaint, conduct independent investigations, and take action accordingly. As in the case of protective behavior, the continuum of more to less 30which questions are designed to identify status- protectiveness, or any of the two other indicators, is explained in the section "The Research Instrument," in the next chapter. 252 position-orientedness ranges from alternatives 1 to 4, Extension of the scope of the hierarchical rela tionship regards those behaviors whereby the organiza tional superior-subordinate relationship is extrapolated beyond the bureaucratic system into nonorganizational spheres of interaction such as social, family, interper sonal, and economic situations. For example, should A and B meet each other at the same party, which of the follow ing behaviors would A expect from B? (1) B should keep ’’the due distance" and not try to join A’s conversational group. (2) B might join A's group, but "not forget that there are differences between a superior and a subordi nate." (3) B should show A some deference, but not behave as formally as if they were in the office. (4) B should treat A independently of their functional relationship. Again, alternative 1 is the most position-oriented behavior of all. On those three indicators— status-protectiveness, emphasis on hierarchy, and extension of the scope of the hierarchical relationship— rests the analysis of position- orientedness of both executives and subordinates, in 253 Brazil and the United Statesto be developed in the next chapter. Functionality and Dvsfunctionalitv of Position-Orientedness in Organiza tional Behavior In the specific field or organizational behavior, the concept of function^ may be approached from several perspectives, one of the most fruitful of which is, undoubtedly, bureaucratic theory. In order to clarify the importance and the uses of the concept in reference to large organizations, some contributions of a few thinkers must be summarized and linked together along with a brief 31por an elaborate discussion of those indicators, see Khan, on. cit.. Chapter VI. •^Nowadays, the concept of function is one of the universals of science. In the social sciences, "function alism1 1 has become the central referent, when not the label, of many schools of thought. Anthropologists like Linton, Malinowski, Herskovits, Radcliffe-Brown, and Ruth Benedict have discussed the functions of cultural patterns exten sively. In psychology, the functionalism of Angell, Dewey, and Carr represented a reaction against the intro- spectionism of Titchner. Economics, particularly after Keynes, has relied more and more on the concept of func tion; possibly, the most manifest indication of this tendency can be found in studies of econometrics. Finally, 254 discussion on the generalities of bureaucratic theory. Social science owes to Talcott Parsons much of the present explanatory power of the concept of function. Approaching it as an analytical tool for the understanding of social systems, Parsons emphasized the usefulness of the concept in providing "criteria of the importance of dynamic factors and processes within the system; . . ."33 and, going further, he called attention to the "other side of the coin," namely, the notion of dysfunction: A process or set of conditions either "contri butes” to the maintenance (or development) of the system or it is "dysfunctional" in that it detracts from the integration, effectiveness, et cetera, of the system. 3“ + structural-functionalism became a distinct sociological school of thought sponsored by men like Parsons, Merton, Levy, and Shills. •^Talcott Parsons, Essays in Sociological Theory (rev. ed.; Glencoe, 111.: The Free Press of Glencoe, 1954), p. 217. Italics in the original. Ibid.. p. 218. Observe that "dysfunctional" still appears between quotes, certainly as an indication of its unusualness or even of its coinage. Nor does the subject index of the book have any entry for "dysfunction" and its cognates. Nevertheless, by lending explicitness to the concept of dysfunction, Parsons was actually open ing new horizons for the theory of bureaucracy. 255 Robert K. Merton not only refined the concepts of function and dysfunction, but also applied them to a species of the whole social system, i.e., bureaucracies. Indeed, in his “Manifest and Latent Functions,"35 Merton explored some important theoretical derivations from those constructs. First, he reinforced the Parsonian argument about dysfunctionality, by explicitly critizing "the tendency to confine sociological observations to the positive contributions of a sociological item to the social or cultural system in which it is implicated,"3^ and by contrasting the two concepts in sharper definition al terms: Functions are those observed consequences which make for the adaptation or adjustment of a given sys tem; and dysfunctions. those observed consequences which lessen the adaptation or adjustment of the sys tem. There is also the empirical possibility of non functional consequences, which are simply irrelevant to the system under consideration.37 oe Robert K. Merton, "Manifest and Latent Func tions," in Social Theory and Social Structure, ed. Robert K. Merton (rev. ed.; Glencoe, 111.: The Free Press of Glencoe, 1963), pp. 19-83. The concept of dysfunction is given two entries in the subject index of the book. 36ibid.. p. 51. Italics in the original. 37ibid. Italics in the original. 256 Second, Merton introduced the ideas of intent ion- ality and cognition so as to distinguish between manifest and latent functions. The former "... are those objec tive consequences contributing to the adjustment or adaptation of the system which are intended and recognized by participants in the system, the latter, "... correlatively, being those which are neither intended nor r e c o g n i z e d ."39 Finally, he used the notion of predic tability as a criterion for differentiating between "anticipated" and "unanticipated" consequences,^9 Merton also deserves the merit of applying such refinements to the analysis of bureaucracy, in order to display not only its functional, anticipated, and intended consequences, but also its dysfunctional, unanticipated, 38Ibid. 39Ibid. ^9In fact, Merton developed a preliminary state ment on this problem in 1938, "The Unanticipated Conse quences of Purposive Social Action," American Sociological Review (1936), pp. 894-904, At that time, however, he did not even mention the word "function," much less "dysfunc tion." 257 and unintended ones.^ It is now time to develop some brief comments on bureaucratic theory, as a background for the discussion of the main theme of this section, function ality and dysfunctionality of position-orientedness in organizational behavior. Bureaucratic theory, as crystallized in Weber's ideal-type,42 may be conceived as both a set of concepts and a set of hypotheses. The first comprise the descrip tive, the latter, the prescriptive, part of the ideal- type.^ Examples of those concepts are specialization, hierarchy, neutrality, or depersonalization, and profes- 4*Robert K. Merton, "Bureaucratic Structure and Personality," in op. cit.. ed. Robert K, Merton, pp. 195- 206. The most extensive and diversified treatment of functions and dysfunctions of bureaucracy in a single piece is: Robert K. Merton et al. (eds.), Reader in Bur eaucracy (Glencoe, 111.: The Free Press of Glencoe, 1952). 42xhe two most known translations of Weber's ideas are: Talcott Parsons (ed.), The Theory of Social and Economic Organization, trans. A. M. Henderson and Talcott Parsons (Glencoe, 111.: Free Press and Falcon's Wing Press, 1947); and H. H. Gerth and C. Wright Mills (eds.), From Max Weber, trans. with an introduction by H. H. Gerth and C. Wright Mills (New York: Oxford University Press, 1958). ^For an insightful treatment of this approach, see: Blau and Scott, op. cit.. particularly, pp. 32-36. A good deal of the ensuing discussion rests on this book. 258 sionalization or career system. Actually, these are some of the characteristics which outline— or “should" outline— the behavioral profile of an "authentic" bureaucracy; they are most useful as analytical tools for organizational research. Examples of the hypotheses^ that make up the prescriptive side of the ideal-type are (1) specialization results in expertness and, ultimately, in efficiency; or (2) depersonalization increases the rationality of decision-making and leads, ultimately, to efficiency. Figure 6 outlines the significant features of the Weberian ideal-type. Observe that all the characteristics, as well as the hypotheses, rest on the same basic assumption: that the maximization of rationality results in the maximiza tion of efficiency. All of them are, therefore, manifes tations of rationality in the search for efficiency. However, as demonstrated in many studies, the ideal-type suffers from what might be termed an "analy tical partiality," in the sense that it provides only a 44These hypotheses were not formulated by Weber as they appear in the examples below. However, their infer ence from his theory of bureaucracy is a commonly accepted matter. Fig. 6.— An outline of the Weberian ideal-type Descriptive Dimension (A set of concepts or characteristics’ ) Specialization A well-defined division of labor, so that each posi tion may be filled by ex perts. Hierarchization A high degree of speciali zation requires coordina tion, the responsibility for which must be well speci fied. This is done through a chain of commands— a hierarchy— whereby offices are organized into a pyra midal structure. Each head is responsible before a superior and exerts author ity over several sub ordinates . Rulification A consistent system of rules governs the bureau crat1 s actions and decisions, so that a disciplined per-;* formance is achieved. Depersonalization Only technical criteria are to orient decisions and actions. Both the external and the internal publics— clients and officials— are to be Prescriptive Dimension (A set of hypotheses) Specialization fosters expertness and, ultimate ly, efficiency. Hierarchization— well- defined responsibility and authority— improves coordination, and, ultimately, efficiency. Rulification leads to operational uniformity, which facilitates coordi nation, and thereby increases efficiency. Depersonalization avoids particularistic judge ments in the interest of individuals rather than of the organization. 260 Fig. 6— Continued Descriptive Dimension Prescriptive Dimension (A set of concepts or (A set of hypotheses) characteristics) Depersonalization (continued) treated irrespectively of per sonal considerations. Sec ondary, rather than primary group criteria, are to orient the bureaucrat's behavior. Meritization Full-time employment oriented Meritization enhances to a lifelong career all of competence and, ulti- which is to be based on the mately, efficiency, professional, rather than personal, qualifications of the individual. 261 functional but misses a dysfunctional analysis of bureau cracy. Merton^"* suggested that the very characteristics designed to increase rationality have both functional and dysfunctional consequences to efficiency. For instance, a well-specified structure of authority may be a leverage for disciplined performance, and adequate identification of responsibility for the application of penalties, on the one hand; and a deterrent to upward communication, risk- taking, and initiative, on the other. Merton's analysis is much in line with Gideon Sjoberg* s concept of "contra dictory functional requirements."46 sjoberg contends that every system presents inherent contradictory functional requirements which lead to the development of contradic tory substructures. An application of this idea to bureaucracy would suggest a chart as the one outlined in Figure 7. The foregoing passages about functionalism and bureaucracy suggest that, if evaluated in terms of its ^^Merton, "Bureaucratic Structure . . . pp. 195' 206. ^See; Gideon Sjoberg, "Contradictory Functional Requirements and Social Systems," The Journal of Conflict Resolution. IV, No. 2 (June, 1960), 198-208. 262 Functional Contradictory to Functional Requirement each other but Requirement ____________________essential to_______; _________________ Depers onalization, E Motivation, identifi- detachment --- -7* f— — cation with organiza F tion. esDrit de corps Conformity to rules, F Risk-taking, initia- disciplined and V 7 t ive, innovat ion, uniform performance f I creativity Standardization, C Flexibility, adjust formalization of --- — adjustive-developmen channels 1 I tal capacity Positional authority E Professional author as the basis of —V ^ . ity as the basis of decision (discipline) f N C Y decision (expertise) Fig. 7.— Contradictory functional requirements of bureaucracy 263 effects upon some given referent, every act of organiza tional behavior may be functional, dysfunctional, or non functional. Both the act and the referent may vary almost infinitely. Moreover, an act which is functional for a given referent may be dysfunctional for another. This is, in brief, the basic mode of analysis of the functional sociologist. The following chapter deals with a particular kind of organizational acts or behaviors— authority relation ships— from the standpoint of their degree of position- orientedness; the specific referent against which the effects of position-orientedness in authority relation ships are to be assessed is the strategy ideology of the instrumental model of change. In other words, position- orientedness is not to be taken as functional or dys functional per se. but insofar as it supports or disrupts that ideology. Now, as demonstrated in Chapter III, the ideology of the instrumental model of change strategy requires, whether implicitly or explicitly, position- oriented authority relationships. The questions to be asked are: What sort of position-orientedness? Is it a "homogeneous,'’ simple, concept? If not, which consequences might result from position-orientedness? The following distinct hypothetical situations provide an introduction to answering those questions: 1. I am the boss here. Mr. X (a member of the public) is to be helped at once; I do not care if others arrived before him. I am the one who sets priorities. 2. 1 am the boss here. Attendance to the public is to be on a first-come, first-served basis. Only in the absolute interest of the service such a criterion may be eventually overlooked. Obviously, both superiors are position-oriented, in that they both rely on formal elements, such as hier archy and legal authority, for the acceptance of their behaviors. But their similarity stops here. One is more oriented towards particularistic, the other towards universalistic, goals. One uses his position for personal rewards, the other does it for organizational accomplish ment. One deviates from bureaucracy, the other conforms to it; one undermines values such as depersonalization, rulificaticn, and meritization; the other reinforces them. In sum, one is functional to, or supportive of, the ideology of the instrumental model of change strategy; the 265 other is dysfunctional to, or disruptive of, that ideology.^ The research design developed in the next chapter centers on a position-orientedness continuum. Such a design, originally developed by Humayun Khan for his com parison of Pakistan and the United States,^8 does not take into account the distinction between functionality and dysfunctionality of position-orientedness. Though aware of the distinction, Khan intendedly left it out of the design of his study. It must be emphasized in all these cases [author ity behaviors] that, throughout this study, there is no intention of showing that one form of behavior is better than another, for. that author says, ^It may be claimed (in fact, it is, by many) that orientations of a universalistic, depersonalized, and technical pattern are essential for large-organizations optimizing their resources. Though the author subscribes such an argument in its broadest sense, the interest of this chapter is confined to the implications of those orientations, or of their antinomies, to the ideology of the instrumental model of change strategy only, with no regard for their relevance for other phenomena (for instance, political development). Such a narrowness of focus seems indispensable for keeping this study under manageable commitments. ^Khan, pp. cit. 266 We are only identifying behaviors, not passing judgement on them.49 However, an analysis of the content of the indica tors of position-orientedness used by Khan and also here shows rather clearly that they are not "neutral"— or, as Merton would put it, indicators of nonfunctional conse quences. Rather, most of them indicate, on the one hand, dysfunctional position-orientedness, and, on the other, functional nonposition-orientedness.50 Such functionality ^Ibid., p. 198, footnote 6. ■^To be more precise, two distinct continua are embraced by and overlap each other in Khan’s indicators. The first continuum ranges from more position-orientedness to less position-orientedness; the second— which is implicit rather explicit— ranges from more dysfunctional to less dysfunctional behaviors. In Khan's study those two continua were usually related in a direct, positive way, i.e., the more position-oriented a behavior, the more dysfunctional it is; correlatively, the less position- oriented, the less dysfunctional a behavior is. We say "usually" because in some items dealing with emphasis on hierarchy position-orientedness may be functional, rather than dysfunctional, if one takes into account the ecology of the bureaucracy where the position-oriented behavior occurs. For example, to be hierarchy-oriented in a bur eaucracy where those in positions of authority tend to be the most competent cannot be considered as dysfunctional. The same would not be true for a bureaucracy where ascrip- tive criteria prevail. The items of the questionnaire which may require a specific interpretation are indivi dually examined in the section "Findings of the Study: Over-all Interpretation," of the next chapter. As of now, the point is that, taken as a whole, the alternative 267 and dysfunctionality are so characterized, of course, in terms of the referent used in this study, the ideology of the instrumental model of change strategy. Indeed, the first of those indicators, status protectiveness, refers to a pattern of authority relationship whereby a superior resorts (or is expected by his subordinates to resort) to acts designed to defend his personal status, irrespective of the consequences of such acts for his organization. This is the case, for instance, when a superior either tries to get rid of a subordinate, or reinforces control over him, because such a subordinate tends to question that superior's orders (no matter, of course, what merits the subordinate's arguments may have). The second indica tor, emphasis on hierarchy, refers to a pattern of author ity relationship which overstresses position at the expense of factors such as competence, skill, open com munication, and group participation. This is the case, for example, of a superior who always delegates to "the next in command" (independently, therefore, of the responses for the items of Khan's questionnaire relate position-orientedness to dysfunctionality, and nonposition orientedness to functionality. 268 competence or specialized knowledge of that subordi nate). The third indicator, more outrightly than the other two, reveals dysfunctional position-orientedness. The very name of it, extension of the scope of the hier archical relationship (to family, social, and socio economic spheres), implies a violence to the bureaucratic ethos. It dispenses with further comments. In summary terms, most of the alternative responses designed to indicate position-orientedness regard behaviors whereby formal elements such as the hierarchy are stressed for the sake of personal rewards and status defense, regardless of thair organizational consequences. It is therefore submitted, that, taken as a whole, the items used in Khan's research and also in the present one are measures of dysfunctional position-orientedness and functional nonposition-orientedness. 5!Las noted before, "the next in command" may usually be the most competent. In such cases, position- orientedness cannot be considered as dysfunctional. Thus, some items of this indicator require a more specific interpretation (see footnote 50, above). Observe that, contrary to emphasis on hierarchy, the other two indica tors imply dysfunctionality, almost by definition. 269 Authority Relationships and Comparative Public Administration One of the most critical issues in the theory and practice of comparative public administration has been the transferability of techniques and analytical tools from one culture to another. Such an issue is prone to remain a headline as long as change agents insist on imposing foreign social inventions— position classification, per formance budgets, and central purchasing, for instance— on client systems whose value textures are often adverse to those inventions. However, though the realities of technical assistance reveal that the technician and the theoretician are frequently far apart from each other— their consensus being but formalistic only— the fact is that comparative public administration has made some impressive accomplishments. It is reasonable to, say that comparative administration— theory, which deals predomin antly with the understanding of administrative phenomena in different settings— is far more advanced than develop ment administration— practice, whose focus is intervention in, or planned change of, the administrative dynamics of 270 different societies. And to the extent that understanding is a prerequisite for rational intervention, development administration can only succeed as it leans on strategies of change emanating from theory and validated by research in comparative administration. From the analyses of trends in the comparative study of public administration and political science as well, one may infer that those disciplines are moving (1) from normative towards more descriptive approaches, (2) from idiographic towards nomothetic efforts, and (3) from structural towards functional p e r s p e c t i v e s .52 jt is suggested that the present study, by focusing on authority relationships from a behavioral approach, aligns with all those tendencies. **^Four of the best known of such analyses are: Ferrel Heady, "Recent Literature on Comparative Public Administration," Administrative Science Quarterly. V (June, 1960), 134-54; Fred W. Riggs, "Trends in the Com parative Study of Public Administration," Intematipna1 Review of Administrative Sciences. XXVII, No. 1 (1962), 9-15; Roy Macridis, The Study of Comparative Government (New York: Random House, Inc., 1955), particularly Chapter I; and Gabriel Almond and James Coleman (eds.), The Politics of Developing Areas (Princeton, N.J.: Princeton University Press, 1960), Chapter I. 271 First, the replacement of an "ought" by an "is" approach reveals, among other things, a consciousness of the need for field research instead of armchair recommen dations, a recognition of the explanatory insufficiency of administrative axioms, and an increasing sensitivity to the distinction between logical and psychological sound ness in administration. The study of authority relation ships in Brazil and the United States, as here proposed, is an example of this new direction. Indeed, it is an effort towards empiricism, towards description rather than prescription, and towards a more sophisticated level of rationality than the logical level only. Second, the search for nomothetic instead of merely idiographic works is suggestive of the need for surpassing the casuistry of the traditional studies, if theory building is to be envigorated; there is no other way of validating or reformulating the existing theory except by the comparative method. It is here where all the efforts towards model building converge; it has become necessary to develop analytical tools which can be applied to several administrative systems concomitantly, so as to enlighten differences and similitudes between such sys tems. In other words, it is indispensable to identify common referents against which various administrative systems can be contrasted. Now, authority relationships are an inherent feature of any formal organization and, therefore, one of the most conspicuous of such common referents. Still more important, authority relationships are a critical variable of the administrative process. By presenting a comparison between the United States and Brazil, the coming chapter, though in a limited way only, is an effort toward nomology in comparative public admin istration; and, to the extent that the strategy of admin istrative reform in other countries of South America is contained in the general lines of the instrumental model, the nomothetic qualities of that study are apparent. Third, the preference for functional perspectives, instead of structural ones, is but a reflection of the current behaviorism of the social sciences. The formal- legalistic studies which would compare, for instance, the structure and functions of the Ministry of Finance in several countries are gradually being overpassed, for it 273 is more and more acknowledged that "similarity in bureau cratic structure does not preclude wide differences in official behavior within this structure."^3 By approach ing the phenomenon of authority from a behavioral per spective, this study lines up with the functionalist trends of comparative administration. It is an attempt towards the understanding of the actual dynamics of authority relationships in the United States and Brazil, with the ultimate purpose of relating those dynamics to the process of administrative change in the latter. It entails, therefore, a functional, rather than structural, approach to the problem. •^Robert V. Presthus, "Behavior and Bureaucracy in Many Cultures," Public Administration Review, XIX, No. 1 (Winter, 1959), 34. CHAPTER V AN ASPECT OF THE BEHAVIORAL PROFILE OF THE BRAZILIAN BUREAUCRACY: AUTHORITY PATTERNS The central purpose of this chapter is to ascertain whether the authority patterns required by the instrumental model of change strategy are those which actually prevail in the Brazilian bureaucracy. As proposed in Chapter IV, the extent to which such requirements are fulfilled by the client system is indicative of the strategic adequacy— and, consequently, of the probabilities of success— of an intended change. An Empirical Study: Rationale Even if the above proposition is accepted, one might wonder about the reasons for examining the actual orientations of Brazilians in their authority relation ships. It would not be enough to respond that any behav ioral study on Brazil is justified, considering the 274 275 current stage of the administrative literature in Portuguese. The question would still remain: Why raise doubts as to the Brazilians' position-orientedness? The Brazilian literature on administration does not help answer such a question. It is almost totally lacking in empirical efforts, the bulk of the works cen tering on either of the two extremes: (1) theoretical discussions which ignore reality, and (2) reports and monographs, often minute and legalistic, which are theory- free. There are very few "linking-pin" studies where some theoretical framework is tested against administrative realities. Indeed, the task of abstracting those reali ties into meaningful constructs— the only way of under standing and generalizing about the Brazilian administra tion— is barely initiated. To this author's knowledge, the only empirical work dealing with authority relation ships in Brazil is by Frank P. Sherwood.* Though highly ^Frank P. Sherwood, "Comparative Authority Orien tations in the United States and Brazil: A Preliminary Probe" (unpublished paper delivered at a Workshop Confer ence for Trainers on New Research Findings in the Behav ioral Sciences held at the University of Southern Califor nia, Civic Center Campus, Los Angeles, April 30, 1964), pp. 1-13. 276 exploratory and tentative in its conclusions, that work is a source for hypotheses about the orientations of Brazil ians towards authority. On the basis of a questionnaire applied to Brazilians and Americans, Sherwood suggests, among other things, that "in crisis situations the bulk of Brazilians say they would follow the man who they regard as most expert, rather than the hierarchical leader. . . At least two important considerations may be derived from this remark. The first is that the hierarchi cal leader is not perceived as combining both expertise or competence and authority. In other words, hierarchy and competence tend not to go hand in hand. The second is that, under normal conditions, the hierarchical leader is followed. However, if the situation reaches some critical point where it becomes necessary to accomplish a given purpose, the hierarchy is relegated to a second order. These patterns of authority relationships bear heavily on any change whose strategy depends upon the chain of command— the hierarchy of authority— for its implementation. It is important, in this case, to identify 2Ibid., pp. 7-8, 277 the motivational basis for the acceptance of the hierarch ical leader. As Sherwood seems to indicate, in the case of Brazil such an acceptance persists only so long as it represents a defense against penalities or actually a retreat from risk-taking. In other words, the hierarchi cal leader is usually followed because, in case of any mistake on the job, "... the wise follower will have his duties spelled out so that he is not the object of attack"3 and will always have a justification for his faults by saying that he was just complying with his superior's instructions. This might be taken as indica tive of a dysfunctional position-oriented pattern of behavior, in which the stress on the formal elements of the organization reflects not a concern with rationality or efficiency, but a retreat from responsibility, which ultimately results in the subordinate's alienation. Initiative and risk-taking, so relevant to the institu tionalization of change, are discouraged; the bureaucrat becomes more and more an expert in insurance against blame. 3Ibid.. p. 12. 278 On the other hand, the majority of Americans answered that they would follow the hierarchical leader even in situations of crisis.4 Contrary to the Brazilian case, this implies (1) a perception of competence and authority as attributes which are combined in the formal leader, and (2) a hierarchical orientation motivated not by defensiveness but by an understanding of the value of the hierarchy for goal accomplishment. This might be taken as indicative of a functional, rather than dysfunc tional, position-oriented pattern of behavior, in which emphasis on hierarchy serves organizational, not personal, objectives. Sherwood further suggests that in Brazil the formal leader is perceived not as a goal-accomplisher for the organization, but as a satisfaction-keeper for the group: "Income for the leader comes, then, through defer ence by the group, personal emoluments, et cetera, and not that which comes from goal accomplishment."** Again, this is an example of dysfunctional position-orientedness, as 4Ibid.. pp. 7-8. 5Ibid.. p. 12. 279 contrasted with the Americans' perception of the formal leader as a goal accomplisher. It must be observed that the questionnaire used by Sherwood did not differentiate between functional and dysfunctional position-orientedness but only between position-orientedness and nonposition-orientedness. How ever, since the instrumental model of change strategy re quires functional position-orientedness in authority relationships, the distinction between functional and dys functional position-oriented behaviors is indispensable to this study.^ In the examples commented upon above, for instance, both Brazilians and Americans were position- oriented, but the position-orientedness of the former was of a dysfunctional, and of the latter of a functional, sort. Other aspects of the behavioral profile of the Brazilian bureaucracy appear to support the hypothesis of a prevalence of dysfunctional behavioral patterns in that ^Actually, from both a theoretical and a pragmatic standpoint, it is possible to distinguish between (1) functional position-orientedness, (2) dysfunctional posi tion-orientedness, (3) functional nonposition-orientedness, and (4) dysfunctional nonposition-orientedness. 280 profile. This is the case, for instance, of patterns of group behavior, which tend to be oriented towards particu laristic, individual-protective, functions. Therefore, this attempt to probe the actual pat terns of authority relationships in Brazil is justified. In a way, the present chapter is both a refinement and a replication of the study by Sherwood. The differences lie primarily in research and statistical design, and in theoretical purposes.? The United States was taken again as the referent for comparison with Brazil. This choice is entirely defensible from a theoretical point of view, for American thought and practice in administration have guided— and one might add, perhaps too uncritically— the Brazilian reform efforts. In other words, the same strategy ideology has been tried in both Brazil and the United States, in spite of their different behavioral Though also exploratory, this chapter draws on a much larger sample and a much more elaborate questionnaire than was the case of Sherwood* s study. As to theoretical purposes, he was primarily concerned with the relationships between level of development and occupational mobility, on the one hand, and position-orientedness, on the other, whereas this chapter is concerned with the implications of position-orientedness for the strategy of planned organizational change. 281 profiles. It can then be hypothesized that, though such a strategy ideology proved to be relatively efficient in the American bureaucracy, the incompatibility between that ideology and the actual behavioral patterns of the Brazilian client system raises serious questions as to the likelihood of success of any change attempt based on such a strategy. It seems only logical to select for compari son the country whence the rationale for and the processes of— and this implies saying, the values of— administrative reform in Brazil have been sought. Hypotheses In Chapter III, it was shown that Brazilian admin istrative reforms have been patterned after an instrumen tal model of change strategy. The strategy ideology of the instrumental model requires (1) a significant degree of functional position-orientedness in executive behavior, and (2) a significant degree of functional position- orientedness in the expectations for executive behavior. Since the purpose of this chapter is to investi gate the extent to which these requirements are fulfilled by the bureaucracy, the hypotheses below were developed directly from the strategy ideology of the instrumental 282 Q model. Thus, the hypotheses governing this chapter are: 1, There is no significant difference in position- orientedness of executive behavior between public adminis trators in Brazil and the United States. 2. There is no significant difference in position- orientedness of expectations for executive behavior between public administrators in Brazil and the United States. The extent to which executives are position- oriented in their behavior is inferred from the column "What Actually Happens?" of the questionnaire. That is to say, how a superior's behavior is perceived by the respond ents is taken as indicative of how executives actually behave in the bureaucracy. Therefore, the testing of ®For the sake of simplicity, the words "functional" and "dysfunctional" are omitted from these hypotheses. However, as noted before, most items of the questionnaire involve position-orientedness of a dysfunctional, and non position-orientedness of a functional, sort. Thus, for the questionnaire as a whole, position-orientedness is taken as dysfunctional, and nonposition-orientedness as functional. For a lengthy discussion on this point, see the section "Functionality and Dysfunctionality of Posi- tion-Orientedness in Organizational Behavior," especially pp. 262-68. The general assumption of position-oriented ness as functional and nonposition-orientedness as dys functional does not apply to few items of the question naire. These items will be individually considered in the interpretation of the data. 283 hypothesis number 1 is based on perceptions of the respondents. Similarly, the extent of position-orientedness in expectations is derived from the column "What Should Happen?" of the questionnaire. The testing of hypothesis number 2 is then based on expectations of the respondents for the behavior of executives in the bureaucracy. The Research Instrument The data were collected by means of a self-admin istered, structured, multiple-choice questionnaire, which g is presented in Appendix A. It consisted of two parts, the first dealing with background information of the respondents, and the latter with position-orientedness in authority relations. The eleven items forming the first part were addressed to the following variables! (1) age, (2) marital status, (3) number of children, (4) occupation, (5) number of years of education, (6) degrees held, (7) field of study, (8) specialized training, and (9) religion. ^Actually, the questionnaire had three parts. Since the last one is not used in this dissertation, it is omitted from Appendix A. 284 The twenty-four items comprising the second part were distributed among the indicators of position- orientedness as follows: 1. Items 1 to 5 dealt with status-protectiveness. 2. Items 6 to 19 dealt with emphasis on hierarchy. 3. Items 20 to 24 dealt with extension of the scope of the hierarchical relationship.^ With the exception of the last four items, which were addressed directly to the respondent, each of the others depicted a hypothetical situation where a superior, A, ha3 available alternative behaviors in response to a subordinate, B. The respondent was asked to provide two answers to each item, one referring to which of those alternative behaviors he actually perceived as being resorted to by A; the other referring to how he, the respondent, expected A to behave. In other words, each item involved both a perception (registered in the column "What Actually Happens?") and an expectation (registered l^xhis numbering order is not the same in the questionnaire given to the respondents. Nor were the questions presented under given headings as they appear in Appendix A. There, for the sake of clarity, the instru ment was rearranged so that the questions fall homogene ously under each indicator. 285 in the column "What Should Happen?") of the respondent. "Expectation" was used in the normative sense, meaning the respondent's opinion as to how A ought to behave. Thus, while perceptions are indicative of behaviors, expectations are indicative of values. The last four items, instead of presenting hypo thetical situations, addressed questions directly to the respondents, and dealt more with attitudes than with behavior. * • * - An instruction sheet accompanied the questionnaire. The respondents were instructed to (1) conceive the hypo thetical situations as occuring in their bureaucratic settings; (2) consider B as subject to some degree of direct supervision by A, thereby eliminating situations wherein B, though subordinate to A, might never see the latter; and (3) to generalize, as much as possible, about the superior-subordinate relationship per se. thereby avoiding specific situations where a number of peculiar factors might operate. 1*-See questions 21 to 24 in Appendix A. 286 Most of the items about position-orientedness were developed by Humayun Khan, on the basis of his personal experience as a public servant in the Pakistani bureau cracy and as a student of public administration in the United States.12 Khan acknowledges the contributions of several studies for the construction of the questionnaire, ". . . notable among which wa3 the preliminary work done by Professor Frank Sherwood and his colleagues of the University of Southern California Public Administration Group in B r a z i l . " ^ The other works cited are by Pownall,^ Hall, O’Leary, and Williams, Berger,^ and •^Mohammed Humayun Khan, "Authority Behavior: An Empirical Approach to Comparative Public Administration" (unpublished doctoral dissertation, University of Southern California, 1965), p. 159. 13Ibid.. pp. 159-60. ^G. A. Pownall, "An Analysis of the Role of the Parole Supervision Officer" (unpublished Ph.D. disserta tion, Department of Sociology, University of Illinois, 1963). l^Jay Hall, Vincent O'Leary, and Martha Williams, "The Decision-Making Grid: A Model of Decision-Making Styles" (unpublished paper of the National Parole Insti tutes, Washington, D.C., 1964). ^■^fforroe Berger, Bureaucracy and Society in Modem Egypt: A Study of the Higher Civil Service (Princeton, N.J.: Princeton University Press, 1957). 287 Peabody.^ The Sample Though it aims at testing the significance of differences between two groups— Brazilians and Americans— this study is to be considered as an exploratory one. Its research methodology is bound by certain limitations which preclude generalizations to either of the two national groups. The sampling procedure is the most serious of these limitations. Indeed, both samples were chosen on a non-probabil ity basis, the choice of subjects being entirely dictated by circumstances. For reasons irrelevant to the content of this chapter, any greater adequacy of the sampling pro cedure would not have been feasible. The Brazilian sample was drawn in approximately equal shares from three states: (1) Rio de Janeiro, (2) Bahia, and (3) Rio Grande do Sul. From a geographical standpoint, those states are located far enough from each ^Robert Lee Peabody, "Authority in Organizations: A Comparative Study" (unpublished Ph. D. dissertation, Department of Political Science, Stanford University, 1960). 288 other so as to assure the total sample the virtue of some differentiation. For example, from a socioeconomic per spective, some stratification can be expected to occur as a result of differences among the three states. An over whelming majority of the individuals were public employees, the few exceptions being business administrators. The presence of the latter is not likely to alter the general characteristics of the total sample at all. The ninety- seven questionnaires which were returned make up the Brazilian sample. The American sample is comprised of 240 indivi duals, 157 of whom are city managers in the State of California, and the remaining 83 are federal, state, and local employees. The two groups are similar to each other in two respects. First, all Americans and almost all Brazilians are public employees. Also, in regard to their occupa tional characteristics, it must be pointed out that they shared some academic training in public administration. In the case of Brazilians, sampling was done at the Brazilian School of Public Administration, in Rio de Janeiro, the Institute of Public Administration, in Rio 289 Grande do Sul, and the School of Administration, in Bahia. Almost all the samplees were public employees undergoing specialized training at the time they filled the question naire. In the case of Americans, the fact of their being city managers generally involved some sort of training in 1 ft public administration. ° Personal Background of the Respondents The respondents' background was the object of the first part of the questionnaire. The samples are, on the whole, significantly different from each other. On only one variable— number of years of occupation— is there no significant difference (p ^>.05). Brazilians are, on the whole, much younger than Americans. This was to be expected, if not for other . reasons, on the basis of the age composition of the popu lations in the two countries. The data on age seem to be ^®For an understanding of the occupational respon sibilities as well as the process of appointment of the California city manager, see: John C. Bo liens, Appointed Executive Local Governments: The Californian Experience (Los Angeles: The Haynes Foundation, 1952); and League of California Cities, Municipal Administrative Assistants (Berkeley: League of California Cities, 1951). 290 reflected in the next variable, marital status. Here, the great majority of Americans (96 per cent) are married, while Brazilians are almost equally split between married (50 per cent) and nonmarried (47 per cent) respondents. In regard to the number of children, the differ ence between the groups is in the "no children” category. More than 50 per cent of Brazilians have no children, as compared to less than 20 per cent in the case of Americans. This is consistent with the youthfulness and marital status of the Brazilian sample. As to the father’s occupation, striking differ ences appear in the comparison of the categories "govern mental employee" and "nongovernmental employee." A high percentage of Brazilians (39 per cent) are sons of govern mental employees, as against only one-third of that per centage (13 per cent) in the case of Americans. Con versely, about 36 per cent of Americans are sons on non governmental employees, as against only 4 per cent of Brazilians. It seems plausible to interpret such differ ences as indicative of the degree of social mobility in the civil service of the two countries, in the sense that in Brasil sons of governmental employees tend to follow 291 their fathers' career more than is the case in the United States. Also, the social origins of the public personnel are more diversified in the latter. The data on number of years of education also con form to what was expected. An overwhelming majority of Brazilians (74 per cent) has no more than, fourteen years of education, as against a minority (20 per cent) in the case of Americans; on the other hand, while no Brazilian has more than seventeen years of education, about 28 per cent of Americans have been in school for eighteen or more years. The data on years of education are consistent with those on degrees held by the respondent, for the percent age of Brazilians who hold no formal degree is about 70 per cent, as contrasted to about 22 per cent in the case of Americans. Brazilian respondents scored impressively higher than Americans in liberal arts (38 per cent to 3 per cent), which shows very well the predominance of that type of education in Brazil. This is reinforced by the fact that no Brazilian is from the field of physical sciences. In the field of public administration, Americans scored much higher than Brazilians (31 per cent to 7 per cent); 292 clearly, this is in a large part a consequence of the com position of the American sample, constituted mostly by city managers. The data on specialized training present some inconsistencies with those on field of study. First, though by a small percentage, more Americans (39 per cent) than Brazilians (33 per cent) have no specialized train ing. Also, more Brazilians (28 per cent) than Americans (22 per cent) have specialized training in public adminis tration; finally, more Brazilians (39 per cent) than Americans (21 per cent) have specialized training in fields other than public administration. Such results are also incongruent with those on years of education and formal degrees held by the respondents. The questionnaire does not allow for any better specification of the responses, for the question was put in too general terms: "Do you have any specialized training or education?" It is plausible, since the responses to the other items are consistent with each other and inconsistent with this one, to imply that this item was answered with quite different 293 interpretations by Brazilians and Americans. Finally, Brazilians and Americans differ signifi cantly in religion. The former are predominantly Catholic (85 per cent to 14 per cent of Americans), the latter pre dominantly Protestant (78 per cent to 5 per cent of Brazilians). These differences between the two samples deserve some final comments. One might argue that factors such as age, education, marital status and other variables of personal background are responsible for the differences in position-orientedness between the two groups. In other words, should the samples be matched on those factors, there might be no difference in the degree of position- orientedness between Americans and Brazilians. Though such an argument is methodologically correct, matching the l^The data on this variable suggest much incon sistency between Americans and Brazilians as to their per ceptions of the hierarchy of degrees. For instance, 2 per cent of Americans and 3 per cent Brazilians appear as holding doctorates. The concept of doctorate is entirely different in Brazil, for in many fields the individual who completes that which corresponds to an American baccalau reate is called "doctor," It seems very suggestive of this possible distortion that no Brazilian holds a Master's degree. two samples would be far from achieving any representa tiveness, for the populations of public employees in Brazil and the United States have different compositions. Let us suppose, for instance, that in Brazil governmental employees average twenty years of age, whereas in the United States they average thirty years of age. In such circumstances, matching the samples as to age would inval idate generalizations about either country. It might then be argued that the samples are representative of their populations just because Brazilians were younger than Americans on the average. Indeed, this author would expect public employees in Brazil to be younger, predom inantly nonmarried, have less children, descend predom inantly from governmental employees, have fewer years of education, hold fewer formal degrees, major predominantly in liberal arts, and be predominantly Catholic, as con trasted with Americans. Data Processing and Statistical Methodology Once the questionnaires were collected, the data were punched on two sets of IBM cards. Coding categories 295 were developed for the two parts of the instrument. 20 The first set of cards, corresponding to the first part of the questionnaire, contained the data on the personal background of the respondents; the other set of cards, corresponding to the second part of the questionnaire, contained all the information relating to position- orientedness. The distributions of the two samples were compared on each item, and chi square used to estimate the probability that they were drawn from the same population. Whenever marginal frequencies in the original dis tributions were too low to provide expected frequencies in all cells of 5 or more, adjacent categories were combined to provide satisfactory expected frequencies.21 Thus, though in the majority of items the continuum from more to less position-orientedness presents four gradations, in some items the collapsing of adjacent categories reduced that continuum to three, or even two, gradations. 20see Appendix B. 21For an explanation of this point, see: Helen M. Walker and Joseph Lew, Statistical Inference (New York: Henry Holt & Co., 1953), pp. 106-108. Status-Protectiveness: Findings 296 This indicator is intended to identify perceptions of, and expectations for, behaviors whereby A, feeling that B is positing a threat to his position, resorts to acts designed to defend that position. Different hypo thetical situations are depicted in the five items which deal with status-protectiveness. Thus, the first item of the questionnaire presents several alternative behaviors to a superior in a situation where one of his subordinates is generally regarded as more competent than he. As in the entire questionnaire, such alternatives are disposed along a position-nonposition-orientedness continuum. The second and third items hypothesize threats to A1s posi tional authority; in item 2, B has the tendency to ques tion A's orders; and in item 3, a specific directive of a superior is questioned by his subordinate. How is A per ceived and expected to behave, protecting his position or acting according to the merits of the objection? Finally, item 4 inquires into the perceptions of, and expectations for, A's behavior when confronted with a specific work situation where his subordinate actually knows more about 297 the problems than A himself; and item 5 involves a super ior’s willingness to assist his subordinate in devel oping a talent which, though absent in A himself, is important for the organization. What Actually Happens: Perceptions The responses on this column, distributed along the position-orientedness continuum, are shown in Table 4. As Table 4 indicates, a significant difference at the 1 per cent level (p <^.01) was found in 3 of the 5 items dealing with this indicator. In item 1, which refers to a situation where B is generally regarded as more competent than A, the difference between Brazilians and Americans would be significant at the 10 per cent level (p <^.10). The results on item 2 show Brazilians and Americans reacting in the same way to a situation where a subordinate tends to question his superior's orders. In regard to the direction of the relationship, Brazilians fell with more frequency than Americans in the most position-oriented cell in all five items. On the TABLE 4 POSITION-ORIENTEDNESS AS INDICATED BY STATUS-PROTECTIVENESS (ITEMS 1 TO 5, PART II) PROPORTIONS OF RESPONSES TO THE QUESTION "WHAT ACTUALLY HAPPENS?" AND RESPONSE CATEGORIES USED IN PARENTHESES Item No. Hypothetical Situation Country Position-Orientedness More Less N 1 B is generally regarded as Brazil .38 (1) .40 (2) .15 (3) .07 (4) 94 more competent than A United States .29 (1) .40 (2) .26 (3) .05 (4) 240 2 B has a tendency to Brazil .51 (1) .49 (2,3,4) 94 question A’s orders United States .45 (I) .55 (2,3,4) 240 3 B does object to a specific ** Brazil .63 (1) .25 (2) .12 (3) 95 directive issued by A United States .35 (1) .48 (2) .17 (3) 240 In an inspection of B's work, ** Brazil .55 (1) .25 (2) .15 (3) .05 (4) 95 4 A finds himself not familiar with it (as he ought to be) United States .37 (1) .09 (2) .16 (3) .38 (4) 239 B has a given talent that A ** Brazil .54 (3) .33 (2) .13 (1) 94 5 lacks. This talent is im portant for the organization United States •15 (3) .58 (2) .27 (1) 212 *Indicates that a difference significant at the 5 per cent level was found. **Indicates that a difference significant at the 1 per cent level was found. t o vo oo 299 22 basis of the sign test such results are highly signifi cant (p .03), supporting the conclusion that, so far as status-protectiveness is concerned, Brazilian public administrators are more position-oriented in their execu tive behavior than Americans. What Should Happen: Expectations The responses on this column are displayed in Table 5. A significant difference was found in 4%of the 5 items concerned with status-protectiveness— items 1, 2, 4, and 5— all at the 1 per cent level (p <^.01). Brazil ians and Americans were not significantly different from each other in their expectations in circumstances where B objects to a specific directive issued by A. An analysis of the direction of the relationship reveals Brazilians as more position-oriented than Americans in 4 out of 5 instances (items 1, 3, 4, and 5). Such results are statistically significant, on the basis of the sign test (p * .03), thereby allowing one to infer 2^For an explanation of this test and a table of its values at given levels of significance, see ibid.. pp. 430-32 and 458, respectively. TABLE 5 POSITION-ORIENTEDNESS AS INDICATED BY STATUS-PROTECTIVENESS (ITEMS I TO 5, PART II) PROPORTIONS OF RESPONSES TO THE QUESTION "WHAT SHOULD HAPPEN?” AND RESPONSE CATEGORIES USED IN PARENTHESES Item Position-Orientedness No. Hypothetical Situation Country ---* More Less N 1 B is generally regarded as ** more competent than A Brazil United States .37 (1,2,3) .27 (1,2,3) .63 (4) .73 (4) 95 240 2 B has a tendency to question ** A's orders Brazil United States .01 (1) .08 (1) .09 (2) • 14 (2) .65 (3) . .67 (3) . 25 (4) 11 (4) 95 239 3 B does object to a specific directive issued by A Brazil United States .52 (1,2) .42 (1,2) .42 (3) .58 (3) 97 240 4 In an inspection of B's work, ** A finds himself not familiar with it (as he ought to be) Brazil United States ,37 (1,2,3) .18 (1,2,3) .63 (4) .82 (4) 97 240 5 B has a given talent that A ** lacks. This talent is im portant for the organization Brazil United States .28 (2,3) .06 (2,3) .72 (1) .94 (1) 96 212 ^Indicates that a difference significant at the 5 per cent level was found. **Indicates that a difference significant at the 1 per cent level was found. u> o o 301 that, so far as status-protectiveness is concerned, Brazilian public administrators are more position-oriented than Americans in their expectations about executive behavior. Emphasis on Hierarchy: Findinas This indicator is intended to identify perceptions of,Hand expectations for, behaviors which reflect a greater emphasis on the position itself as against factors like competence, group morale, skill, and human relations. Such a behavioral pattern was inquired into by fourteen items of the questionnaire, namely, from items 6 to 19. Several hypothetical situations are depicted in these fourteen items. Every situation admits a number of alter natives, ranging from most to least position-oriented behaviors. Items 6, 7, and 8 deal with the patterns of com munication in an authority relationship. Item 6 presents a situation where the subordinate addresses his superior by the letter's first name; item 7 focuses on A's manner— directive or suggestive— of issuing orders; and item 8 investigates the importance attached to formal display of 302 deference, by positing a situation where A arrives to work and, walking into his office, finds B there. How defer entially would A expect B to react? Items 9 and 10 relate to the dynamics of group participation and group decision-making; both items depict situations where A is faced with several alternative behaviors when presiding at a meeting of his subordinates or a conference which is also attended by specialists. How will he behave? Will he overlook technical competence and specialization in order to reassert his hierarchical seniority? Item 11 probes into A's style of delegation. Position-orientedness can be inferred, for instance, if the superior's criterion for delegating work is the hier archy— in military terms, "the next man in command"— or other factors like competence, more familiarity with the specific task to be delegated, or personal confidence on the subordinate. A hypothetical situation involving such alternatives is presented in this item. Item 12 involves a personal conflict between two subordinates who are themselves unequal in rank. The extent to which the superior attaches importance to the 303 hierarchy, rather than to the merit of the case per se. in the solution of this problem gives an indication of the degree of position-orientedness in his behavior. Item 13 inquires into the superior's tendency to keep the administrative process within the "proper channels," even if at the expense of a faster dispatch of documents. The situation depicted in this item involves attendance of the public. Item 14 refers to the pattern of interpersonal relationships between a superior and his subordinates. A measure of position-orientedness can be obtained from the extent to which those relationships are patterned more by factors like rank and seniority than by others like com petence or personal empathy. Item 15 posits another conflict between two sub ordinates who are themselves unequal in rank. This time, however, the conflict is of a technical, rather than per sonal, nature. The two subordinates diverge from each other as to recommendations on a decision to be taken. On what criteria will A, as a decision-maker, mostly base his solution: rank of the subordinates, their specialization, the merit of the two recommendations independently of 304 their signers, or the judgements of other members of the work group? The remaining items dealing with emphasis on hierarchy— 16 to 19--aim at identifying the predominant basis of authority (hierarchical position? expertise? interpersonal skillfulness?) in the client system. The respondent is addressed directly in these four items and asked about his perceptions of the basis of authority as well as his expectations for the way such an authority ought to be exercised. What Actually Happens: Perceptions The responses on this column, distributed along the position-orientedness continuum, are shown in Table 6. This indicator revealed striking differences be tween the two groups. Indeed, with the exception of two cases (items 12 and 18), a significant difference was found in all the others--items 6 to 11, 13, 14, 16, and 19 (p. .01); and item 15 (p -<^.05). Therefore, in 11 out of 13 items Brazilians and Americans are significantly different from each other. TABLE 6 POSITION-ORIENTEDNESS AS INDICATED BY EMPHASIS ON HIERARCHY (ITEMS 6 TO 19, PART II) PROPORTIONS OF RESPONSES TO THE QUESTION “WHAT ACTUALLY HAPPENS?" AND RESPONSE CATEGORIES USED IN PARENTHESES Item Posi tion-Orientedness No. Hypothetical Situation Country - More Less N 6 B addresses A by the latter's ** Brazil .39 (1) .20 (2) .32 (3) .09 (4) 95 first name United States .06 (1) .17 (2) .37 (3) .40 (4) 207 A wants B to do a given work. ** Brazil .59 (2) .41 (1) 96 7 What will be A's manner of issuing orders? United States .29 (2) .71 (1) 212 A arrives at work and meets ** Brazil .42 (I) .12 (2) • 14 (3) .32 (4) 94 8 B. How deferential would the first expect the latter to be? United States .07 (1) .21 (2) .25 (3) .47 (4) 211 9 A presides at a meeting of ** Brazil .42 (3) .41 (1) .17 (2) 94 his subordinates United States .19 (3) .52 (1) .29 (2) 240 A presides at a conference ** Brazil .66 (3) .19 (1) .15 (2) 94 10 which is also attended by several specialists United States .35 (3) .51 (1) • 14 (2) 238 11 A delegates some work ** Brazil .32 (1) .10 (2) .58 (3) 94 United States .44 (1) .34 (2) .28 (3) 240 A faces a personal conflict Brazil .19 (4,3) .48 (2) .32 (1) 93 12 between two subordinates who are themselves unequal in rank United States .30 (4,3) .39 (2) .31 <i) 238 u> o Ul TABLE 6*“Continued Item Position-Orientedness No, Hypothetical Situation Country — More Less ---^ N 13 While attending the public, ** A is solicited to do a task that is generally done by B Brazil United States United States .26 (2) .50 (2) .22 (1) .05 (1) .52 (3) .45 (3) 94 239 14 A interacts with his B's. ** What factors pattern this interaction mostly? Brazil United States .60 (1) .27 (1) .11 (2) .33 (2) .08 (3) .15 (3) •21 (4) .25 (4) 95 210 15 A faces a technical conflict* between two subordinates who are themselves unequal in rank Brazil United States .47 (1) .58 (1) .30 (2,3) .18 (2,3) .23 (4) .24 (4) 94 212 16 Respondent's perception of ** the basis of his own author- ity Brazil United States .74 (1) .31 (1) .08 (2) .18 (2) .18 (3) .51 (3) 92 233 17a Respondent's perception of ** the uses of his own authority Brazil United States .51 <2) .04 (2) .34 (I) .77 (1) .15 (3) .19 (3) 91 231 18 Respondent's perception of the basis of his superior's authority Brazil United States .71 (1) .63 (1) .11 (2) .17 (2) .18 (3) .20 (3) 92 202 19 Respondent's perception of ** the uses of his superior's authority Brazil United States .54 (2) .22 (2) .22 (1) .43 (1) .24 (3) .35 (3) 91 225 ®The alternative answers to this question are not mutually exclusive. Because of its logical inconsistency, this item is eliminated for all purposes. ♦Indicates that a difference significant at the 5 per cent level was found. **Indicates that a difference significant at the 1 per cent level was found. 306 307 As far as the direction of the relationship is concerned, Brazilians scored with a higher frequency than Americans on the most position-oriented cell. The former were more position-oriented in 9, and the latter in only 4, of the 13 items dealing with perceptions of emphasis on hierarchy. Submitted to a sign test, such results are statistically significant (p = .046). It is worth point ing out that the three items which presented the widest differences between the two groups refer to situations in which the most purely formal aspects of an authority rela tionship are involved. This is the case in item 6, which deals with the manner of addressing a superior; item 8, which refers to a superior's emphasis on deferential be havior by his subordinates; and item 4, which focuses on the factors that pattern the interaction between A and B. Needless to say, these items refer to kinds of position- oriented behaviors which are dysfunctional to the strategy of the instrumental model. In summary, the results on perceptions of emphasis on hierarchy support the conclusion that Brazilian public administrators are more position-oriented in their execu tive behavior than Americans. 308 What Should Happen: Expectations The responses on this column are displayed in Table 7. As pictured in that table, a significant differ ence was found in 7 of the 13 items dealing with emphasis on hierarchy. Observe, also, that, as in the case of status-protectiveness, the difference between the two groups tends to be less apparent when one deals with expectations. This seems to be only natural, since individuals tend to be more critical of their answers about "what should happen" than about "what actually happens." The reduction of the sensitivity of the items to discriminate between the expectations of Brazilians and Americans may be indicated by the need for collapsing adjacent categories. Thus, in only two cases (items 6 and 8) were the calculations done on the basis of four cate gories, whereas in four instances (items 11, 13, 14, and 15) the calculations had to be done on the basis of a dichotomy, for the marginal frequencies of the original distributions of such cases were too low to provide sat isfactory expected frequencies. TABLE 7 POSITION-ORIENTEDNESS AS INDICATED BY EMPHASIS ON HIERARCHY (ITEMS 6 TO 19, PART II) PROPORTIONS OF RESPONSES TO THE QUESTION "WHAT SHOULD HAPPEN?" AND RESPONSE CATEGORIES USED IN PARENTHESES Item Position-OrientedneBs No. Hypothetical Situation Country — More Less - — N 6 B addresses A by the letter's ** Brazil .19 (1) .12 (2) .33 (3) .36 (4) 96 first name United States .07 (1) .23 (2) .18 (3) .52 (4) 206 A wants B to do a given work. Brazil .89 (2) .10 (1) 97 7 What will be A's manner of issuing orders? United States .95 (2) .05 (1) 212 A arrives at work and meets B.** Brazil ,22 (1) .24 (2) .34 (3) .20 (4) 96 8 How deferential will the first expect the latter to be? United States .04 (1) .28 (2) .32 (3) .36 (4) 211 9 A presides at a meeting of Brazil .0.^ (3) .36 (1) .55 (2) 97 his subordinates United States .0& (3) .36 (1) .58 (2) 238 A presides at a conference ** Brazil .05 (3) .83 (1) .12 (2) 97 10 which is also attended by several specialists United States .25 (3) .50 (1) .25 (2) 239 11 A delegates some work * Brazil .23 (1) .77 (2,3) 96 ________________________________________________ _United States .44 (1) .56 (2,3)_________________________ 239_ A faces a personal conflict Brazil .58 (4,3) .12 (2) .30 (I) 96 12 between two subordinates who are themselves unequal in rank United States .64 (4,3) .15 (2) .21 (1) 238 While attending the public, Brazil .92 (2) .08 (1,3) 95 13 A is solicited to do a task that Is generally done by B United States .93 (2) .07 (1,3) 238 A interacts with his B’s. * Brazil .74 (1,2) .26 (3,4) 97 14 What factors pattern this ______ interaction mostly? United States .80 (1,2) .20 (3,4)_________________________ 209 309 TABLE 7“-Continued Item No. Hypothetical Situation Country Position-Orientedness More Less---^ N 15 A faces a technical conflict ** between two subordinates who are themselves unequal in rank Brazil United States .13 (1,2) .33 (1,2) .87 .67 (3.4) (3.4) 97 210 16 Respondent's perception of the basis of his own Brazil .13 (1) .24 (2) •63 (3) 95 authority United States .22 (1) .26 (2) .52 (3) 228 17a Respondent’s perception of the uses of his own Brazil .02 (2) .79 (1) .19 (3) 95 authority United States .00 (2) .82 (1) .18 (3) 226 18 Respondent's perception of the basis of his superior’s Brazil •31 (1) •31 (2) .38 (3) 90 authority United States .44 (1) .27 (2) .29 (3) 194 19 Respondent's perception of ** the uses of his superior's Brazil .38 (2) .32 (1) .30 (3) 94 authority United States .U (2) .51 (1) .39 (3) 217 aTlie alternative answers to this question are not mutually exclusive. Because of its logical inconsistency, this item is eliminated for all purposes. *Indicates that a difference significant at the 5 per cent level was found. **Indicates that a difference significant at the 1 per cent level was found. 311 Contrary to the previous instances, the direction of the relationship here shows Americans as more position- oriented than Brazilians in 9 (7, 10, 11, 12, 13, 14, 15, 16, and 18) out of 13 items. Submitted to a sign test, such results are statistically significant (p<^.046). On the basis of this indicator, one must reject the null hypothesis that there is no significant differ ence in position-orientedness between the expectations of Brazilian and American public administrators about execu tive behavior. Even though only a slight majority of the items (7 out of 13) presented significant differences between the two groups, the decision to reject the null hypothesis is further supported by the statistical test of the direction of the relationship, on the basis of which it is concluded that Americans are more position-oriented than Brazilians in their expectations for executive behavior. Extension of the Scope of the Hierarchical Relationship: Findings This indicator aims at identifying perceptions of, and expectations for, behaviors whereby the organizational 312 superior-subordinate relationship is extrapolated beyond the bureaucratic system into nonorganizational spheres of interaction such as social, family, interpersonal, and economic situations. It is suggested that behaviors extending the scope of the hierarchical relationship are indicative of position-orientedness. Thus, items 20 to 24 deal with the carrying of the organizational superior-subordinate relationship into other spheres of life; in item 20, a social situation is hypothesized; item 21 focuses on a family situation; item 22 addresses an interpersonal relationship; finally, both items 23 and 24 are concerned with socioeconomic situations. What Actually Happens: Perceptions The responses on this column, distributed along the position-orientedness continuum, are shown in Table 8. The chi square test revealed that in 4 (items 20, 21, 22, and 24) of the 5 items dealing with extension of the scope of the hierarchical relationship, a difference significant at the 1 per cent level (p *01) was found. TABLE 8 POSITION-ORIENTEDNESS AS INDICATED BY EXTENSION OF THE SCOPE OF THE HIERARCHICAL RELATIONSHIP (ITEMS 20 TO 24, PART II)— PROPORTIONS OF RESPONSES TO THE QUESTION "WHAT ACTUALLY HAPPENS?" AND RESPONSE CATEGORIES USED IN PARENTHESES Item Position-Qrientedness No. Hypothetical Situation Country ^ -- More Less V N iiu. 20 nypoiuc l l v o i oxuuaLiou A and B meet at the same ** Brazil .17 (1) .49 (2) iiCOO — .27 (3) .05 (4) El 95 party United States .04 (1) .20 (2) .56 (3) .20 (4) 211 Behavior carrying the official** Brazil .56 <1) • 23 (3) .20 (2) 96 21 relationship into family situations United States .34 (1) .19 (3) .4? (2) 211 Behavior requiring personal ♦♦ Brazil .22 (1) .34 (2) •37 (3) .06 (4) 95 22 work to be done by subor dinates United States .02 (1) .08 (2) .60 (3) .30 (4) 211 23 B owns a much bigger and more Brazil .32 (1) .50 (3) .18 (2) 95 expensive motor car than A United StateB .22 (1) .61 (3) .17 (2) 212 B moves in social circles that** Brazil .29 (1) .50 (3) .21 (2) 94 24 are rated higher than those in which A moves United States .42 (1) •17 (3) .41 (2) 212 ♦Indicates that a difference significant at the 5 per cent level was found. ♦♦Indicates that a difference significant at the 1 per cent level was found. 314 It is worth pointing out that this indicator is the one most unequivocally related to dysfunctional position- orientedness in organizational behavior. In regard to the direction of the relationship, in four cases, also (items 20, 21, 22, and 23), Brazilians fell into the most position-oriented category with a higher score than Americans. On the basis of a sign test, such results are statistically significant (p * .03). The two items in which the widest differences were found refer to situations where a superior is depicted asking his sub ordinates to do personal chores for him (item 22) and carrying the official relationship into family affairs (item 21). These are the most clear examples of personal ization of the hierarchical relationship, a mode of be havior entirely antagonistic to the requirements of the instrumental model. In summary, the results on perceptions of exten sion of the scope of the hierarchical relationship support the conclusion that Brazilian public administrators are more position-oriented in their executive behavior than Americans. 315 What Should Happen: Expectations The responses on this column are displayed in Table 9. As can be seen, a difference significant at the 1 per cent level (p .01) was found in 3 of the 4 items dealing with this indicator. As in the other two indica tors , the differences regarding expectations are much smaller than those on perceptions, which again may be indicated by the need for collapsing adjacent categories. In 3 of the 4 items on which calculations were made, the chi square test was based on dichotomies. In respect to the direction of the relationship, Brazilians appear as more position-oriented than Americans in 2 items (items 21 and 22), while the opposite direction was found in 2 items also (items 20 and 23). On the basis of a sign test, these results are not significant (p = .50). Thus, though it showed that the two groups differ significantly from each other, this indicator failed to provide empirical grounds for a conclusion about the direction of the relationship. TABLE 9 POSITION-ORIENTEDNESS AS INDICATED BY EXTENSION OF THE SCOPE OF THE HIERARCHICAL RELATIONSHIP (ITEMS 20 TO 24, PART II)— PROPORTIONS OF RESPONSES TO THE QUESTION "WHAT SHOULD HAPPEN?" AND RESPONSE CATEGORIES USED IN PARENTHESES Item Position-Orientedness No. Hvnothetical Situation Country ^— More Less — 5 > N 20 A and B meet at the same party Brazil •47 (1,2,3) .53 (4) 97 United States .49 (1,2,3) .51 (4) 212 Behavior carrying the official** Brazil .13 (1) .53 (3) .34 (2) 96 21 relationship into family situations United States .09 (1) .24 (3) .67 (2) 210 Behavior requiring personal ** Brazil .49 (1,2,3) .51 (4) 96 22 work to be done by subor dinates United States .26 (1,2,3) .74 (4) 210 23 B owns a much bigger and more ** Brazil .16 (1,3) .84 (2) 97 expensive motor car than A United States .33 (1,3) .67 (2) 211 B moves in social circles that Brazil .01 (1) .05 (3) .93 (2) 97 24s are rated higher than those in which A moves United States .02 (1) .02 (3) .96 (2) 212 ^Indicates that a difference significant at the 5 per cent level was found. **Indicates that a difference significant at the 1 per cent level was found. aNo calculations could be made on the basis of this distribution. The data fail to discriminate between the two groups. 317 Findings of the Study: Over-all In te rare tat ion In the three preceding sections, the data on each indicator were analyzed separately. For the purpose of making some generalizations and taking decisions about the hypotheses under test, the partial results for each of the three indicators of position-orientedness are now brought together and interpreted as a whole. Perceptions The data on perceptions strongly support the con clusion that Brazilian and American public administrators differ significantly from each other in their executive behavior. Indeed, a significant difference was found in 18 of the 23 items of the questionnaire. In the first indicator, status protectiveness, the differences were significant in 3 of the 5 items; in the second indicator, emphasis on hierarchy, 11 of the 13 items showed signifi cant differences; and in the third indicator, extension of the scope of the hierarchical relationship, the differ ences were significant in 4 of the 5 items. 318 Since it is essential to the rationale of this study to ascertain the relative concentration of the groups along the position-orientedness continuum, an analysis of the direction of the relationship between Brazilians and Americans was made for every item of the questionnaire. The results on this part show Brazilians as more position-oriented than Americans in 18 of the 23 items, a proportion which is statistically significant at the 1 per cent level (p <^.005) on the basis of a sign test. Brazilians were more position-oriented than Americans in all 5 items dealing with status-protective- ness, in 9 of the 13 items dealing with emphasis on hierarchy, and in 4 of the 5 items concerning extension of the scope of the hierarchical relationship. On the basis of the above results, the first hypothesis must be rejected. In other words, the execu tive behavior of Brazilian public administrators tends to be characterized by dysfunctional position-orientedness. Insofar as the instrumental model of change strategy requires the prevalence of an exactly opposite pattern of behavior— one which is characterized by functional position-orientedness— the implications of this finding 319 for the reform attempts are of primary importance. The analysis of those implications is developed fully in the last chapter of this dissertation. Here, the purpose is only to bring some empirical evidence on the incompatibil ity between the behavioral requirements of the instrumental model and the behaviors that actually shape the profile of the client system, the Brazilian bureaucracy. Expectations The data on expectations did not discriminate as sharply between the two groups as in the case of percep tions. This is clearly indicated by the need to collapse adjacent categories of original distributions, so as to have satisfactory expected frequencies for the chi square test. Indeed, in 8 (items 1, 4, 5, 11, 14, 15, 22, and 23) of the items which presented significant differences between the two groups, it became necessary to collapse the position-orientedness continuum into two cells only. By contrast, in the case of perceptions, it was not neces sary to dichotomize any single one of the eighteen items in which the two groups differed significantly from each other. 320 Though the null hypothesis is rejected— for the data provide considerable evidence that the expectations of Brazilians and Americans about executive behavior are significantly different from each other— the critical finding here, from the standpoint of change strategy, is that the conflict between the characteristics of the Brazilian bureaucracy and the strategy ideology of the instrumental model is much more acute in regard to behav iors (perceptions) than it is with respect to values (expectations). This suggests the existence of potential forces favorable to change which are to be considered in devising an appropriate strategy, a point which will be explored later. This lack of clarity and consistency in the dif ferences between the values of Brazilians and Americans is also revealed by the analysis of the direction df the relationships between the two groups. Thus, all items included, Brazilians were more position-oriented in ten (items 1, 3, 4, 5, 6, 8, 9, 19, 21, and 22), and Americans in twelve (items 2, 7, 10, 11, 12, 13, 14, 15, 16, 18, 20, and 23) instances. On the basis of a sign test, such a proportion is not significant (p = .58). Again, when one 321 deals only with the fourteen items on which the two groups differed significantly from each other, there is an almost even split between Brazilians and Americans. The former were more position-oriented in 8 (items 1, 4, 5, 6, 8, 19, 21, and 22), and the latter in 6 (items 2, 10, 11, 14, 15, and 23), of those 14 items. On the basis of a sign test, such a proportion is very likely to occur by chance ( p > . 2 1 ) . However, one should not infer from the above re sults that the two groups tend to be indifferently con centrated on the several gradations of the position- orientedness continuum. Indeed, this would correspond to saying that, so far as position-oriented expectations are concerned, the groups do not differ from each other, a conclusion that the data do not support at all. Hence the need for a more refined analysis of the items on which Americans held expectations which were more position- oriented than those of Brazilians, so as to check on whether position-orientedness, as indicated by such items, might have different consequences in the ecologies of the Brazilian and the American bureaucracies. That is to say, it is necessary to discover whether given position- 322 oriented behaviors might be dysfunctional in one setting and functional in another. Such an analysis is justifi able on at least three grounds. First, it is a well accepted point in social science that the same stimulus is bound to elicit differ ent responses— and, therefore, generate distinct conse quences— under different circumstances. Thus, emphasis on hierarchy may produce functional consequences in a bureau cracy where those holding positions of authority are usually the most competent among the personnel. The same could not be said of a situation where occupancy of a position of command is based on ascription, rather than achievement. Second, though the assumption was made that position-orientedness (as indicated by the items of the questionnaire) tends to be of a dysfunctional type, while nonposition-orientedness (or, evidently, less position- orientedness) tends to represent functional behaviors, such an assumption is only valid for the questionnaire as a whole. It is worth pointing out, for instance, that two of the three Indicators of position-orientedness— status- protectiveness and extension of the scope of the 323 hierarchical relationship— are clearly dysfunctional , as far as the ideology of the instrumental model is concerned. Indeed, such indicators regard behaviors oriented toward personal, at the expense of organizational, goals. Brazil ians were significantly more position-oriented in status- protectiveness (p <^.031), and no significant differences were found in extension of the scope of the hierarchical relationship (p * .50). Taken together, these two indica tors showed Brazilians as more position-oriented than Americans. However, the third indicator, emphasis on hier archy, cannot be taken as dysfunctional in all cases. For example, position-orientedness is likely to elicit oppo site consequences in items 6 and 15. In the first case, B addresses A by the latter*s first name, an informality which is objected to by A. In item 15, however, A sup ports B in the case of a conflict on technical matters between B and C. Also, in terms of the ideology of the instrumental model (here entirely supported by bureaucra tic theory), the hierarchy must be used to pursue the 324 organization's goals;23 and in administrative practice, the hierarchy may be, and is often, used to the benefit of the organization. Hence the need for qualifying which items might indicate functional, and which ones might imply dysfunctional, expectations. The third reason for developing a more specific analysis is that only in regard to emphasis on hierarchy were expectations of Americans more position-oriented than those of Brazilians. This becomes even more meaningful if one considers that, in the case of perceptions as well, 4 of the 5 items on which Americans were more position- oriented than Brazilians deal with the same indicator, emphasis on hierarchy. Hence the question: Why only in regard to some items concerning emphasis on hierarchy have both the perceptions and expectations of Americans de viated from the general patterns shown in the data? The expectations of Americans were more position- oriented than those of Brazilians in 9 of the 13 items 2^observe that, almost by definition, the same ideology (and bureaucratic theory as well) would proscribe behaviors seeking status-protectiveness or extending the scope of the hierarchical relationship, regardless of possible functional by-products of such behaviors. 325 dealing with emphasis on hierarchy. Not all of such 9 items lend themselves to a qualification on the basis of the rationale to be developed below. However, for seven of them (items 11, 12, 13, 14, 15, 16, and 17), the argu ment that position-orientedness is indicative of func tional behaviors in the United States and dysfunctional ones in Brazil seems quite plausible. In the case of two items only (items 7 and 10), such an argument is inappli cable, a difficulty that may well be attributed to the imperfections of the instrument. In items 11, 12, and 13, Americans were more position-oriented than Brazilians, in regard to both per ceptions and expectations. Item 11 depicts a situation in which A has to delegate some work. The most position- oriented alternative is one where the delegation goes to the next man below A in the hierarchy, whereas in the least position-oriented alternative A delegates the work to a subordinate whom he learnt to trust the most because of a personal relationship. Clearly, here position- orientedness must be taken as functional, and nonposition- orientedness, by contrast, as dysfunctional, particularly if one accepts the assumption that, in general, ascendancy 326 to positions of authority in the United States depends more on technical competence than is the case in Brazil. In item 12, A faces a personal conflict between two subordinates who are themselves unequal in rank, B and C. In this hypothetical situation, C sends a com-.; plaint about his superior B to A, without first bringing it to B's notice (a behavior which is formally proscribed in both the United States and Brazil). The four alterna tives, ranging from more to less position-orientedness are as follows: (1) A would penalize C for not bringing it to B's notice first; (2) A would return it to C and instruct him to bring the complaint to B's notice first; (3) A would refer the complaint to B for his comments; and (4) A would examine the complaint, conduct independent investigations, and take action accordingly. It is sub mitted that, of all four alternatives, only the second may be considered functional. Indeed, alternative 1 seems to overstress the hierarchical value, without making a constructive use of the conflict. To be sure, B might even ignore the procedures to file complaints against his colleagues or superiors. His penalization would not build up discipline in the organization. The consequences of 327 alternative 3 would probably result in conflict between B and C, and between A and C as well. And the consequences of alternative 4 would be, at best, the disregard for authority and hierarchy, with the stimulation of further similar acts. Contrariwise, alternative 2 would represent a developmental experience for C, besides reinforcing respect for organizational authority. Since both Tables 6 and 7 present the data on item 12 already collapsed for the purposes of the chi square test, it becomes necessary to present the raw data on this item. Thus, in regard to perceptions, the data on item 12 were originally distributed in the following pro portions : 1 2 3 4 Brazilians .10 .09 .48 .32 Americans .02 .28 .39 .31 As can be seen, Americans were more position- oriented than Brazilians only in the functional position- oriented alternative, 2. The same occurred in the case of expectations, as the original distributions of the data show: 328 1 2 3 4 Brazilians .06 .52 .12 .30 Americans .02 .62 .15 .21 According to item 13, while attending the public, A is solicited to do a routine task generally done by B. The least position-oriented alternative is clearly dys functional ,24 whereas the most position-oriented one is functional.^5 As shown in Tables 6 and 7, Brazilians scored higher than Americans in the dysfunctional cate gories in both perceptions and expectations. In regard to item 14, it is suggested that the first and last alternatives are the most dysfunctional 26 of all. Brazilians scored higher than Americans on those alternatives, in reference to perceptions, as can be seen in Table 6. As to expectations, the data were originally distributed in the following proportions: 2^See Appendix A, item 13, alternative 3. 25ibid., alternative 2. ^Ibid.. item 14, alternatives 1 and 4, 329 1 2 3 4 Brazilians .06 .68 .16 .10 Americans .08 .73 .17 .02 Though Americans scored higher in the first altern ative, the difference is almost nonexistent, whereas the last alternative— the most dysfunctional of all and also the one where the widest differences between the two groups occur— shows a higher proportion of Brazilians than Americans. In item 15, Americans were more position-oriented than Brazilians as to both perceptions and expectations. The situation depicts A facing a conflict on technical matters between two subordinates who are themselves un equal in rank, B and C. It seems crucial to consider the ecological differences between the Brazilian and American bureaucracies in order to interpret the options of the respondents to this item. As noted before, in a bureau cracy like the American, where those in positions of com mand tend to have the necessary competence to perform their duties, the compatible or functional behavior is to support the hierarchy. The same behavior may be highly 330 dysfunctional in a bureaucracy where personal, not tech nical, criteria prevail, as in the case of Brazil. Thus, the fact that Americans were more position-oriented must be taken as indicative of functional behaviors (percep tions) and values (expectations), whereas the lesser position-orientedness of Brazilians may be both functional or dysfunctional, depending upon the specifics of the situation. Similar reasoning is applicable to item 16, which presents Brazilian respondents perceiving the basis of their own authority as residing in their hierarchical position but not their competence,27 and expecting to be accepted by their subordinates mostly on the basis of good personal relations with them.28 Contrariwise, Americans expect to gain the acceptance of their subordinates on the basis of both rank differences and professional competence, and actually stress the latter more.2^ Therefore, the position-orientedness of Americans is functional, whereas 22See Appendix A, item 16; and Table 6. 28See Table 7. 2^Ibid. 331 the nonposition-orientedness of Brazilians is dysfunc tional. Finally, item 18 deals with the respondents' per ception of the basis of their superior's authority, and the responses of the two groups are distributed exactly as those in item 16, therefore eliminating the need for further comments. The foregoing analysis of the individual items dealing with emphasis on hierarchy allows for a conclusion that could not have been taken from a purely statistical standpoint. It is therefore submitted that American and Brazilian public administrators not only differ signifi cantly from each other in their expectations about execu tive behavior, but such difference tends to be in the direction of functional position-orientedness and also functional nonposition-orientedness, in the case of Americans; and of dysfunctional position-orientedness and also dysfunctional nonposition-orientedness, in the case of Brazilians. 332 Summary and Conclusions This chapter attempted to probe the degree of position-orientedness that might pattern authority rela tionships in Brazil. The United States was chosen as the point for comparison. Three indicators of position- orientedness were used, namely: (1) status-protectiveness, (2) emphasis on hierarchy, and (3) extension of the scope of the hierarchical relationship. For each of those indicators, a twofold analysis was developed, one part dealing with perceptions and the other with expectations. Perceptions were taken as indicative of the actual behavior of executives; expectations were used in the normative sense of how an executive ought to behave. Therefore, expectations represent values about executive behavior. Two hypotheses governed this research. They were developed directly from behavioral requirements that a client system must fulfill in order for a change based on the instrumental model to succeed. In regard to authority patterns, such requirements are that there is a predominance of a functional position-orientedness in both executive behavior and expectations for executive 333 behavior. Therefore, the hypotheses herein tested were: 1. There is no significant difference in position-orientedness of executive behavior between public administrators in Brazil and the United States. 2. There is no significant difference in position- orientedness of expectations for executive behavior be tween public administrators in Brazil and the United States. On the basis of the content of the alternative responses for the items of the questionnaire, position- orientedness was considered as indicative of dysfunctional perceptions and expectations; conversely, nonposition- orientedness was considered as indicative of functional perceptions and expectations. Though such an assumption was valid for the questionnaire as a whole, it was not tenable in regard to some items dealing with emphasis on hierarchy. A closer analysis of the alternative responses for some items revealed that position-orientedness was related to functional, and nonposition-orientedness to dysfunctional, behaviors and values. Hypothesis number 1 was tested on the basis of data regarding perceptions, and hypothesis number 2 on 334 the basis of data about expectations. Both hypotheses were rejected, for the data supported the conclusion that the groups differ significantly from each other as to both perceptions and expectations. It must be pointed out that the differences were much greater in the case of perceptions, or behaviors, than of expectations, or values. As to the direction of the relationship between the two groups, the general pattern is one of Brazilians scoring higher than Americans in dysfunctional perceptions and expectations, both of a positional and of a nonpositional type. Put into the broader context of this dissertation, the findings of the present chapter suggest that the pre vailing pattern of executive behavior in the Brazilian bureaucracy conflicts with the requirements of the strategy of change that has guided the attempts to reform public administration in that country. However, while the incon gruence between the characteristics of the client system and the requirements of the instrumental model is rather sharp in regard to behaviors, it is much less apparent in the case of values. In other words, dysfunctional behaviors (both positional and nonpositional) are not 335 as much desired as they are practiced. Therefore, it may be suggested that, to the extent that expectations, or values, which reject dysfunctional patterns of executive behavior come to prevail in the client system, the poten tial for change will be increased in the bureaucracy. From the standpoint of change strategy, this is a crucial element to be explored. As a matter of fact, this element has been overlooked in the Brazilian experience of admin istrative reform, since the model of change strategy which has prevailed in that experience— the instrumental model— is cognitively, rather than valuationally, oriented. Brazilian change agents have not identified so as to explore, from a strategic standpoint, those expectations which would be supportive of change efforts. The implications of the above points for both past and future reform attempts are discussed in the final chapter. Here, the purpose was only to depict the salient features of the behavioral profile of the Brazilian bur eaucracy in what concerns authority patterns. CHAPTER VI ANOTHER ASPECT OF THE BEHAVIORAL PROFILE OF THE BRAZILIAN BUREAUCRACY: TEMPORAL ORIENTATIONS The main purpose of this chapter is to probe the prevailing temporal orientations of the Brazilian bureau cracy, with a view towards relating such orientations to the requirements of the instrumental model of change strategy. The selection of this specific trait--temporal orientations— for the study of the behavioral profile of the Brazilian bureaucracy may be rationalized on at least three grounds: (1) the relevance of time in conditioning human and, consequently, bureaucratic behavior; (2) the essentiality of the temporal dimension as an element of the strategy ideology of the instrumental model; and (3) the astonishing scarcity of both theoretical and empirical works dealing with the influence of temporal orientations 336 337 on organizational behavior,*’ A brief discussion of the first two of the above points may well serve as an intro duction to the empirical study which is carried out in this chapter. Time and Human Behavior: A Conditioning Relationship The phenomenon of time is inherent, and conse quently, universal, to human life. It is also one of the few common denominators of cultures, whatever their age or stage may be. Both man and society are historical systems, o in that they move across a time cycle from birth to death. *-Obviously, the literature on time and motion studies is abundant. Those, however, are not laid on behavioral rationales at all. Thus, the need persists for relating time and organization from a behavioral perspec tive, as it is attempted here. ^The forms of cultural birth and death vary a great deal and are a controversial matter among culturologists. The classical contribution on the rise and death of cul tures is Spengler's cyclical theory whereby all cultures pass through the same stages of birth, adolescence, maturity, decline, and death, much like the phases of human life. See: Oswald Spengler, The Decline of the West (New York: Alfred A. Knopf, Inc., 1926). A com parable but much less deterministic view is Arnold J. Toynbee's "challenge and response" theory. See: Arnold J. Toynbee, A Study of HiBtorv. abridged by D. C. Somervell (New York: Oxford University Press, 1946). 338 Here stands the first aspect of the conditioning of human conduct by the tensoral dimension: man being mortal, time becomes a boundary condition, for its ultimate scarcity. As pointed out by Sorokin: Each of these activities [of man] requires some time for its performance. Since the total quantita tive time alloted to human life is limited, no individual can perform at any moment of his existence all the potentially possible activities with their motives. These crowd one another and compete for the time necessary to carry them on.^ As a consequence of its ultimate scarcity, time exerts another sort of restraint on human behavior; it requires the establishment of priorities, so that, under given temporal limitations, more valued goals can be pursued at the expense of less valued ones. Thus, time scarcity requires human behavior to be orderly or sequen tial, synchronized or coordinated, and expeditious or fast,^ in order for goal-achievement to be optimized. ^Pitirim A. Sorokin, Sociocultural Causality. Space. Time (New York: Russel & Russel, Inc., 1964), p. 209. ^The concepts of sequence, synchronization, and rate, as used in this chapter, were developed by: Wilbert £. Moore, Man. Time and Society (New York: John Wiley & Sons, 1963). These concepts will be expanded upon later. 339 Indeed, it seems a moot point that, “other things being equal," time is best utilized if behavior is sequen tial, synchronized, and fast. This is exactly the assump tion underlying the time and motion studies developed by pioneers like Taylor and the Gilbreths. It was necessary to eliminate duplications and reduce the number of steps of routines, thereby increasing the speed of the work; it was indispensable to rectify workflows so that step 3 actually precedes, not follows, step 4, thereby imposing a sequence to the performance of a task; finally, it was necessary to set up routines in such a way as to avoid bottlenecks at any given step because of dependence on inputs from other workflows, thereby synchronizing the work. As it was argued in Chapter III, the same sort of rationale oriented the strategy of the Brazilian reforms, notably those in the fields of 0 and M, which we called functional reforms, and budgeting. The above set of assumptions is subject to at least two criticisms. First, it is devoided of any behav ioral foundation; it is confined to the instrumental or physiological aspect of the temporal dimension. Second, it fails to perceive that, in reality, "other things" are 340 not equal; that . . the perceptions and symbolism that may trigger a set of behaviors in a particular culture will produce quite different responses in another,for "the perceived . . . [world] is the world that determines behavior."6 In sum, it overlooks the fact that time is perceived in different, and sometimes conflicting, ways, and that such different perceptions forge quite different behaviors of men in society and in organizations. It is to this particular aspect of the conditioning of human conduct by one’s temporal orientations— a behavioral view, therefore— that we now turn. The notion that man is, to a large extent, a pro duct of his total environment is commonplace in modem social science. Available evidence on this matter may be found in thousands of contributions based on a culture- and-personality approach. Another truism today is the idea that cultures differ from each other as to patterns of behavior, attitudes, symbolism, social sanctioning, ^Frank P. Sherwood, "The Clock and the Specialized R and D Society," unpublished paper, 1965, 1965, p. 1. (Dittoed.) ^Harold J. Leavitt, Managerial Psychology (2d ed.; Chicago: University of Chicago Press, 1964), p. 170. 341 value hierarchy, and so on, "Cultural differences" has been both a used and abused expression. One does not need to belabor the above notions to infer from them that different cultures, or different sub cultures within the same culture, breed different indivi duals, different primary and secondary groups, and differ ent organizations. Hence the virtues of an ecological perspective to the study of social behavior. The contention that time, as a cultural variable, conveys different messages and thereby elicits different behaviors is demonstrable in several ways. Mythical or not, the British punctuality is well known around the world. "Would you give me two little minutes (minutinhos?" (minutinhos)?" is a most preferred Brazilian preface to a half-hour long talk. The Sioux Indians had no word for time; the notions of lateness and futurity did not exist in their life space.^ In Latin America, ". . . where time is treated rather cavalierly, one commonly hears the expression, 'Our time or your time?' 'Hora americana. hora 7 Edward T, Hall, The Silent Language (Greenwich, Conn.: Fawcett Publications, 1965), p. 25. First pub lished in hard cover by Doubleday & Co., Inc., 1959. 342 g melicana?1 I r The option for one or another implies dif ferent degrees of punctuality. The concept of full-time employment, for instance, is not the same in Brazil as in the United States, being much less strict in the latter than in the former. In short, "like other cultural phenomena, temporal frames of reference vary profoundly from society to society. The effects of such variations on human behavior cannot be denied. Temporal orientations centered on the past, for instance, are less favorable to the change pro cess than those projected toward the future. In cultures where time is not perceived as a scarce commodity, wage systems based on temporal units may not be economically viable. The operation of an assembly line requires some minimum concern for synchronization and sequence. To be sure, "... all human conduct (and probably all organic behavior) is conditioned bv the time perspectives of the — ■" ■ ' ■■ ' ■ 1 — — h i in ' — ■■ 8Ibid.. p. 19. ^Irving Hallowell, "Temporal Orientation in Western Civilization and in a Pre-Literate Society," The American Anthropologist. XXXIX (October, 1937), 647. individual and of his culture. Therefore, the concept of time may be applied to the study of many aspects of human conduct in different settings. This chapter is concerned (1) with the condi tioning of an individual's behavior by his time perspec tives in a particular setting, large-scale organizations or bureaucracies, and (2) with the implications of such a conditioning relationship for the process of organiza tional change. On the basis of the rationale outlined up to this point, we suggested that (1) organizations are, to a great extent, cultural products; (2) cultures differ from each other, to a greater or lesser degree, in regard to several variables, including temporal orientations, and (3) man's behavior is largely a function of his temporal orientations. Therefore, organizations with distinct pre vailing temporal orientations are likely to emit different responses to the same stimulus. That is to say, stimuli that have led to functional consequences in one bureau cracy may bring about dysfunctional, or simply nonfunc- ■^Lawrence K. Frank, "Time Perspectives," Journal of Social Philosophy. IV, No. 4 (July, 1939), 294. Italics added. 344 tional, consequences, in another. Though quite obvious a point, this has profound implications for planned organi sational change, particularly in regard to strategies used by change agents to sensitize a given client system. The following section examines the temporal orientations which are required by the instrumental model of change strategy. The Instrumental Model and the Temporal Prerequisites of Its Strategy .Ideology The significance of time for social change hardly needs any elaboration. In the most authoritative opinion of Wilbert Moore: "The sense of time and the perception of change are inextricably linked in human experience. The link works both ways; neither time nor change is a dependent variable.This also applies to the more specific field of planned organizational change. Since perceptions of, and expectations about, time exert a pro found influence on human behavior, and since planned organizational change depends a great deal upon the pre- ^Wilbert E. Moore, Social Change (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1963), p. 22. 345 vailing behavioral patterns in the client system, it follows that the temporal dimension becomes a critical variable for the deliberate inducement of change. From the standpoint of change strategy, then, two kinds of information are indispensable to an assessment of the strategic adequacy of any given change attempt: (1) which time-related behaviors the strategy requires from the client system, and (2) which characteristics actually prevail in the client system, in regard to time. The present section deals with the first, and the remainder of this chapter with the latter, of those kinds of informa tion. The task of identifying which temporal orienta tions the instrumental model of change strategy requires from a client system can only rest on a deductive method of inference, at least at the present stage of theoretical development of the matter. To be otherwise, empirical studies of applications of the instrumental model to several client systems with different temporal orientations would have to be available. By identifying relationships between the outcomes of several change attempts (success or failure, on the basis 346 of any given criterion) and the prevailing temporal orientations of the client systems where such attempts had been made, those studies would provide an empirical basis from which one could induce the set of temporal orienta tions of the client system that conditions the viability of a reform patterned after the instrumental model of change strategy. However, such studies do not exist. As said be fore, the temporal dimension of organizational behavior is still almost unexplored. Also, the instrumental model of change strategy, as a theoretical construct, is a product of this study; therefore, it could not have been put to use in social research before. Though parts of that model, often under different labels or definitions, have been employed in administrative studies (the most common examples are studies dealing with Weber’s bureaucratic model), the instrumental model as such— a set of strategic characteristics of a change attempt— has never been explored.^ p0r all these reasons, the identification of ^The reader is reminded of three points. First, the instrumental model is not to be equated to a bureau cratic model of change strategy, even though they have 347 the prerequisites of the instrumental model as to the temporal orientations of the client system is arrived to deductively. Notwithstanding the limitations of a deductive approach from the perspective of scientific methodology, it is possible to arrive to a logically sound (though empirically untested) rationale from which to infer the characteristics of the instrumental model. This rationale centers around three referents: (1) Moore's synchroniza tion- sequence -rate trilogy, (2) some strategic character istics of the Brazilian administrative reforms, and (3) some theoretical and empirical contributions of the literature on formal organizations. Moore's Trilogy; Synchronization. Sequence. and Rate In his Man. Time and Society. Wilbert E. Moore develops the concepts of synchronization, sequence, and several common elements (Chapter I, pp. 27-28). Second, the focus of the instrumental model is on the implications of its characteristics for the fate of a change attempt from the standpoint of change strategy. Third, strategic implications are not the focus usually taken by studies which explore Weber's bureaucratic model. 348 rate and defends their usefulness as differentiators of time-ordering styles of social systems— from man to society— in their responses to time inperatives. Synchronization implies simultaneity of actions. When such actions— as is the case in bureaucracies— must not only be simultaneous but also reciprocally functional, so as to accomplish some given purpose, a coordinative re quirement is implicit in the concept of synchronization.^ Now, in order to achieve coordination, it is necessary to schedule. Therefore, synchronization as a goal depends ultimately upon scheduling as a process, as a behavioral commitment. Since the need for synchronization, and consequently, for scheduling, is ". . . most readily per ceived where individual or undisciplined and intermitent effort is plainly inadequate for the accomplishment of some task,1'15 and since the strategy of reform in Brazil required disciplined and continuous performance, one may 13 JMoore, Man. Time and Society, p. 8. ^Simultaneous but reciprocally nonfunctional, or dysfunctional, actions (example: an acrobat failing to catch his partner in the air, though both jumped simul taneously) cannot be considered as synchronized. ^^Moore, Man. Time and Society, p. 46. 349 logically derive the conclusion that scheduling, both as a value and as a behavior, constitutes one of the temporal prerequisites of the instrumental model. Sequence involves several interrelated but inter- distinguishable notions, the most conspicuous of which are order and priority. The first is more related to the mechanical succession of events in a prescribed way; the latter is more suggestive of the valuational roots of a system's (man or society) propensities to establish pre ferential criteria for the allocation of its time. Actually, order and priority, as above defined, are only different perspectives of the same phenomenon. In the words of Moore: The sequential ordering of activities essentially provides a priority schedule in the strict sense, which mav reflect priorities in the loose sense of relative values. Thus the admonition, "work before play," provides a rank order as well as a temporal order of actions.16 This study uses the concept of sequence from the perspective of priority criteria whereby time is allocated in a given system. An attempt is made to link the notion of time priorities to the Weberian notion of bureaucratic ^Ibid.. p. 48. Italics added. 350 impersonalization. Thus, Moore's "work before play" is rephrased in the terminology of the bureaucratic model as "the organization before the individual." In other words, the bureaucratic model (here equatable with the instru mental model) requires an individual's working hours to be devoted to the pursuing of organizational, rather than personal, objectives. As demonstrated in Chapter III, the Brazilian reforming agency, the DASP, lacked qualified personnel in sufficient number to control the allocations of time by individuals throughout the bureaucracy. This limitation became more serious after the fall of the dictatorship, when democratic guarantees ruled out the compliance power that the reformers had enjoyed. There fore, and more than ever, the reform effort required that the prevailing orientations in regard to the allocation of time in the client system were "naturally" organizationally- oriented, rather than personally-oriented. In short, an organizational orientation in the use of time constitutes another temporal prerequisite of the instrumental model. Before passing on to the third of Moore's concepts, rate, a brief comment on another aspect of sequence is worthwhile. It regards Hall's notion of monochronism, 351 i.e., a one-thing-at-a-time orientation.^ For Hall, monochronism is one of the most distinctive features of the sequence-ordering patterns of Americans in their use of time. Applying the concept to the bureaucratic milieu, Sherwood observes: The United States culture is also characterized by a Mone thing at a time" norm, a sequence commitment of rather significant meaning to administration. Since our satisfactions have tended to derive from goal accomplishment rather than process activity itself, we have the need to stick with something until it is com pleted. Interruptions and discontinuity of activity have rather consistently been regarded as bad; indeed, one of the principal points of the United States Government's famous program in work simplification during World War II was based on the assumption that people are more efficient when they work in a few, functionally related areas. When related to some of the strategic characteris tics of the administrative reform movement in Brazil— intermitency, process orientation, discontinuity— such remarks provide grounds for comparison between the ecologies of the American and Brazilian bureaucracies, as 17Hall, op. cit.. pp. 138-39. The concept of monochronism is not used in this study at all. However, it is submitted that it contributes to the understanding of the administrative reform process in Brazil, and adds to the theoretical content of this work as well. *-®Sherwood, o p . cit.. p. 7. 352 well as the strategies which guided their reform attempts. Observe, for instance, that while the strategy of reform in Brazil has been wholistic, in the United States it has been gradualistic. Thus, in the American experience, there has occurred a steady, continuous succession of goals, "one at a time." After given goals are satisfac torily attained, others, increasingly broader, are adopted. Contrariwise, the wholism of the instrumental model of change strategy implies, almost by definition, palychronism, in the sense of "everything at a time." Indeed, the Brazilian reform movement required an army of Jack-of-all-trades; since the change agents were not sufficient to constitute even a regiment, the consequences were formalism, intermitency, and, worse than all, goal displacement, rather than goal succession. Rate involves a relationship between the frequency of occurrence of events, on the one hand, and a given unit of time, on the other. The importance of rate— as a matter of fact, of synchronization and sequence as well— ultimately derives from timefs inherent scarcity. ''When an order of events must be fitted into a more or less pre cise time period, the temporal order must include an 353 19 activity rate." It seems unnecessary to elaborate on the expeditiousness that the Brazilian reforms required from the individuals in the bureaucracy. The establish ment of deadlines, often so unrealistic, is enough evidence of the need for a considerable concern with fast ness of action, for a "sense of urgency"^ on the part of the client system. These orientations, however, can only occur if, and to the extent to which, time is perceived as a scarce resource, a point on which one may find wide differences between cultures, and, therefore, between bureaucracies. In the United States, for instance: . . . The perception of time as a scarce resource is perhaps no more apparent than among administrative officials. They have generally internalized the goals of their organizations, recognize that mission accomplishment will depend on collective effort, and therefore feel the pressure of time to make their influence felt as widely as possible. As a result, many books on administration are filled with sugges tions about better time use; vet they seem to miss the point because sequence [and rate as well, one may add] is a function of a total value system, not lust a mechanistic question of scheduling. ■^Mbore, Man. Time and Society, p. 50. 20Hall, op. cit., pp. 136-37 and 139. 2*-Sherwood, op. cit.. p. 8. Italics added. 354 The question to be asked, then, in relation to the present study is: To which extent is time perceived as a scarce commodity in the Brazilian bureaucracy? This ques- tion underlies one of the hypotheses of the empirical comparison between the temporal orientations of Brazilian and American public administrators, carried out in the next section. Two final remarks about the concept of rate are worth making. First, rate may represent the most critical element of Moore's trilogy, as far as change strategy is concerned. This is so because, "... where change is more valued than permanence, rate becomes instrumental in heightening the concern about time as a scarce resource."22 And second, to the extent that rate is highly valued but geared to personal, rather than organizational, ends, the process of bureaucratization is bound to collapse. This possibility is a consequence of the ultimate scarcity of time for both the individual and the organization. In deed, . the scarcity of organizational time and the 22Ibid. 355 23 scarcity of individual time intersect," and the accommo dation of this conflict may well be on a priority (or sequence-ordering) basis whereby the organizational time is entirely subordinated to the individual’s own inter ests. Temporal Exigencies of Centralization and Who1ism One of the most salient features of the reform movement in Brazil was centralization of the housekeeping activities for the whole bureaucracy into a single agency, the DASP. Prior to the reforms, those activities were independently performed by each ministry; there did not exist a system, in the sense of a well integrated set with some characteristics common to all its parts; there was no interdependence between those parts. Indeed, one of the major purposes of the reforms was to give the public bureaucracy the properties of a system. A necessary con sequence of this— for which one would find support in systems theory— was a great need for coordinative ability of both the central agency and the field ones. It seems 23Moore, Man. Time and Society, p. 10. 356 dispensable to belabor the point that coordination implies synchronization and, ultimately, scheduling. Whenever the coordinative ability is a prerequisite for goal achieve ment, it becomes evident that scheduling is an equally sine qua non condition. Besides scheduling, expeditious ness is also a temporal value the need for which increases directly with centralization. Indeed, the greater the centralization, the greater and more critical the inter dependence between parts, and the greater the need for rhythm (often, of an increasingly accelerated rhythm). As suggested before, expeditiousness, as a behavioral commit ment, depends directly on the valuation of time as a scarce commodity. The temporal requirements that resulted from the centralization approach of the reforms were made more compulsive because of another characteristic of the change strategy that has prevailed in the Brazilian experience, namely, wholism. Since the reformers were only a handful of experts in administration, a small elite, and since the reforms involved the whole bureaucracy at once, it became impossible to have delegates of the central agency, with a common technical background and value orientations 357 (principally about the reforms themselves) to perform the role of change agents in the field. As pointed out before, the forces of change never came to form a constellation within the context of the Brazilian bureaucracy; rather, they gravitated in, and around, one solitary star, the DASP. This increased the dependency of the central agency not only on the scheduling ability and prompt action of the field agencies, but equally on the prevalence of an organizational, rather than personal, criterion of prior ity in the individuals' allocation of their working hours.^ Particularly in regard to this last requirement— organizational orientation— the dependence of the central .?> agency on the field units came to a climax with the fall of the dictatorship. Indeed, if administrative imperson- alization was to some extent fostered by the use, or at least the potential availability, of a compliance model, 2^This organizational versus personal orientation in time patterns may be related to the notions of func tionality and dysfunctionality of position-orientedness in authority patterns. Interrelating these notions may lead to the identification of consistencies between different traits of the behavioral profile of a bureaucracy. It may be hypothesized, for instance, that dysfunctional author ity patterns are indicative of personal orientation in time allocation. it was not so under democratic conditions. These two characteristics of the change strategy that has oriented the Brazilian administrative reforms— centralization and wholism— suggest the need for schedul ing , expeditiousness, and organizational orientation in whichever context the strategy is applied. However, when the specific characteristics of the Brazilian administra tive ecology are considered--reduced administrative cap ability, a reforming agency, the DASP, aloof within the bureaucracy, and the abrupt replacement of a dictatorship by democratic rules— those requirements become a matter of survival for any reform based on such a strategy. To the extent that the client system does not fulfill them, reforms based on a centralization-wholism approach are likely to be bottlenecked. To redesign a routine and demonstrate the rationality of the proposed change may not be enough to achieve actual change. In other words, as long as the rationality of the intended change does not find echo in the valuational orientations and behaviors which prevail in the client system, the materialization of change is not likely to occur. As is often the case, a wide gap may separate the abstract (rational) from the concrete (behavioral). 359 Supports from the Literature on Large Organizations The temperal prerequisites of the instrumental model may also be inferred from several theoretical and empirical pieces of the literature on large organizations, particularly from those on bureaucracy.^ Essentially, bureaucracy implies an optimizing value system, with its attendant implications for temporal patterns. In Merton's words: "The chief merit of bureaucracy is its technical efficiency, with a premium placed on precision, speed, expert control, continuity, discretion, and optimal returns on output.Clearly, precision implies synchro nization and, therefore, scheduling; speed is just a dif ferent form of saying expeditiousness, which itself re flects a perception of time as a scarce commodity; and 25por this matter, the instrumental model may be taken as similar to a bureaucratic model of change strategy. 26Robert K. Merton, "Bureaucratic Structure and Personality," in Social Theory and Social Structure, ed. Robert K. Merton (rev. ed.; Glencoe, 111.: The Free Press of Glencoe, 1963), p. 196. Italics added. 360 continuity implies a long-run perspective of time, a temporal orientation that is less common than it appears at first. In an extrapolation of his findings on bureaucra tic development in England, France, Brandenburg-Prussia, the American colonies, and the United States, Carl Fried rich provides some grounds for inferences about the temporal prerequisites of the instrumental model. He generalizes: Six elements or aspects recur in a developing bureaucracy in demonstrable institutionalization; centralization of control and supervision (hierarchi cal aspects), differentiation of functions, qualifica tion for office, objectivity, precision and contin uity. and secrecy. The first three of these aspects are organizational, the latter are b e h a v i o r a l .27 Friedrich* s generalization is doubly significant for the present study: (1) because he points up to the temporal element, and (2) because he approaches time patterns as a behavioral prerequisite for the process of bureaucratization. Therefore, his remark is in line with 2^Carl J. Friedrich, "Some Observations on Weber's Analysis of Bureaucracy," in Reader in Bureaucracy, ed. Robert K. Merton et al. (Glencoe, 111.: The Free Press of Glencoe, 1952), p. 29. Italics added. 361 the rationale of this dissertation, according to which an instrumental strategy of change requires given behav ioral patterns— among them, those related to time orienta tions— to prevail in the client system. From a strategic standpoint, the question to be asked, then, is: What if only the organizational aspects (see quotation from Friedrich, above) are sought, as if the behavioral ones would logically follow? Which are the likely results of a "partial" bureaucratization process— one which is con fined to the organizational aspects— in a client system whose behavioral profile is patterned by behaviors which are antagonistic to those required by that process? The temporal prerequisites of bureaucracy were laid down by Weber himself. In his discussion of the technical advantages of bureaucracy, he stated: Prec is ion. speed. unambiguity, knowledge of the files, continuity, discretion, unity, strict sub ordination, reduction of friction and of material and personal costs— these are raised to the optimum point in the strictly bureaucratic administration, and especially in its monocratic form.28 ^^Max Weber, "Bureaucracy," in From Max Weber: Essavs in Sociology, ed. H. H. Gerth and C. Wright Mills (New York: Oxford University Press, 1959), p. 214. 362 Again, precision implies synchronization, which rests on an attitudinal commitment to scheduling; speed means expeditiousness, which requires that time be per ceived as a scarce commodity; and continuity implies a long-run perspective of time, which may be related to con cern for scheduling. The bureaucratic model both seeks and requires the optimization of these time patterns for the sake of organizational, not personal, achievements. Therefore, if Weber had labelled the temporal exigencies that his model makes to the client system, he might well have called them "an optimizing-organizational pattern of temporal orientations." Such a label is submitted as a synthesizer of the temporal prerequisites of the instru mental model of change strategy. Qptimizing-Oreanizational Temporal Orientations In view of the multiplicity of time-related notions spinned out in this section— among them, synchro nization, scheduling, precision, perception of time as a scarce commodity, sequence, order, speed, rate, expedi tiousness, long-run versus short-run time perspective, sense of urgency, monochronism, polichronism, and 363 continuity— and also because the concept of optimizing- organizational temporal orientations may embrace all of those notions, it becomes necessary to the understanding of the forthcoming study (1) to identify which aspects of temporality will be included in the concept of optimizing- organizational temporal orientations, and (2) to relate such aspects to the concepts presented in this section. As operationalized in Chapter I, optimizing- organizational temporal orientations refer to time pat terns which are characterized by a high concern for scheduling and expeditiousness, and by the prevalence of organizational, as against personal, goals in the alloca tion of working hours, in the client system. This concept is an attempt to achieve a theoretical integration of Moore's synchronization-sequence-rate trilogy, the strategic patterns of the Brazilian experience of admin istrative reform, and the temporal exigencies of Weber's bureaucratic model. The first of Moore's concepts, synchronization, is represented in the instrumental model by concern for 29See Chapter I, p. 46. 364 scheduling. As suggested before, synchronization, as a result, depends directly on scheduling, as a behavioral commitment. By similar reason, concern for scheduling also represents Weber's notion of continuity. The third of Moore’s concepts, rate, is repre sented in the instrumental model by perception of time as a scarce commodity. The plausibility of such a connection derives from the fact that rate of change, in a given system, is meaningful only in view of time constraints. In other words, only if time is scarce— if it is per ceived as scarce— will rate tend to be accelerated. By the same reason, perception of time as a scarce commodity also encompasses Weber1s notions of speed and expeditious ness, and Hall's sense of urgency. Finally, the second of Moore's concepts, sequence or order, is represented in the instrumental model by organizational versus personal orientation in one's allocation of his working hours. Thus, Moore's sequence is interpreted in terms of the priority criteria that pre vail in the client system for solving the inevitable con flict between the scarcity of organizational time and the 365 scarcity of individual time.^® Such an interpretation provides a linkage between Moore’s sequence and Weber's impersonalization. Observe that this part comprises the second half of the "optimizing-organizational" dyad, whereas the other concepts constitute the first half. Temporal Orientations in the Brazilian Bureaucracy: An Empirical Study As suggested before, the greater the compatibility between the requirements of a strategy of change as to the behavioral patterns of the client system, on the one hand; and the actual prevailing behavioral patterns of that system, on the other, the greater the probabilities of 31 success of the intended change. The degree of such a compatibility, by indicating the probabilities of success of a change effort, may also be taken as indicative of the adequacy of the strategy used in the change attempt. Since the main concern of this dissertation is with the adequacy of the prevailing strategy of adminis- 30Moore, Man. Time and Society, p. 10. 3ISee Chapter IV, pp. 234-36. 366 trative reform in Brazil, and since that strategy— as demonstrated in the last section— depends on given temporal patterns on the part of the client system, it follows that, in order to assess the adequacy of the strategy, one needs to identify and describe which time patterns actually pre vail in the system. This seems to be enough to justify an empirical study of the temporal orientations of the Brazilian bureaucracy. Hypotheses The hypotheses below were directly developed from the requirements of the instrumental model of change strategy as to the prevailing time patterns of the Brazil ian bureaucracy. As discussed before, those requirements are (1) perception of time as a scarce commodity, (2) con cern for scheduling, and (3) organizational orientation in the individuals' allocation of their working hours. Therefore, the hypotheses governing this chapter are: 1. There is no significant difference between public administrators in Brazil and the United States in regard to their perception of time as a scarce commodity. 2. There is no significant difference between public administrators in Brazil and the United States in 367 regard to their concern for scheduling. 3. There is no significant difference between public administrators in Brazil and the United States in regard to their use of work hours for organizational, as against personal, purposes. The Research Instrument The data were collected by means of a self administered, structured questionnaire, which is presented 3? in Appendix C. It consisted of four parts, the first dealing with background information about the respondents, the second with perception of time as a scarce commodity, the third with concern for scheduling, and the fourth with 32xhe twenty-eight items which comprise the research instrument are part of a more extensive question naire developed by Mujib A. Sheikh, for the purpose of his doctoral dissertation, at the University of Southern California. Sheikh's study also deals with temporal patterns, but it rests on a different theoretical frame work and does not involve Brazil. This author translated Sheikh's questionnaire from English into Portuguese. Data on Brazil were specifically collected for this disserta tion in late 1965. Data on the United States were collected by Sheikh in the middle of 1965. This author is indebted to his colleague for the use of both the items of the questionnaire and the data on the United States. 368 personal versus organizational orientation in the alloca tion of work hours.^3 The first part, involving four questions, was addressed to the following variables: (1) nationality, (2) age, (3) supervision experience, and (4) length of supervision experience. The remaining twenty-five questions were so dis tributed among the temporal dimensions under analysis: 1. Questions 5 to 13 dealt with perception of time as a scarce commodity. 2. Questions 14 to 23 dealt with concern for scheduling. 3. Questions 24 to 28 dealt with personal versus organizational orientation in the allocation of work hours. Except for the questions on background information 33 The numbering order in which the items of the instrument are presented in Appendix C is not the same as in the questionnaire given to the respondents. Nor were the questions presented under given headings as they appear in that appendix. There, for the sake of clarity, the instrument was rearranged so that the questions fall homogeneously under each one of the dimensions under analysis. 369 and of question number 8,^ all the others refer to either perceptions or expectations of the respondent about temporal patterns. As in the study of authority patterns, perceptions were taken as indicative of actual behaviors, whereas expectations--used in the normative sense— were taken as values. The Sample Both Americans and Brazilians were chosen on a nonprobability basis. In sampling terminology, the two groups constitute accidental samples.^5 Practical reasons eliminated any possibility of a greater accuracy of the sampling process. In view of such a limitation, it is methodologi cally indefensible to develop generalizations for either of the groups. However, it is legitimate to consider the data as preliminary evidence about the two countries, on 34"What is the length of the official work week for government employees in your country?" ^-*For an elaboration on sampling methodology and its implications upon the generalizability of findings, see: Claire Selltiz et al.. Research Methods in Social Relations (New York: Holt, Rinehart & Winston, 1962), Appendix B, pp. 509-45. 370 an exploratory basis. The Brazilian sample is comprised of 115 indi viduals, 40 of whom are federal employees, 71 are state employees, 1 is a local employee, and 3 were not identi fied. Out of the 40 federal employees, 23 work in the State of Bahia, and the other 17 in the State of Guanabara. Therefore, although this dissertation is confined to the federal bureaucracy, the Brazilian sample is predomin antly a state-oriented sample.^ State employees come from eight different states in Brazil, as follows: (1) Rio Grande do Sul— 27; (2) Bahia— 19; (3) Minas Gerais— 11; (4) Pernambuco— 5; (5) Amazonas— 3; (6) Ceara^--3; (7) Paraiba— 1; (8) Parana^-1; (9) state unidentified— 1. With the exception of those from Rio Grande do Sul and Bahia, the other state employees were sampled in Rio de Janeiro, at the Brazilian School of Public Administra tion. This is the reason for the relatively wide diver sity of states represented in the sample. From both a •^These were limitations which the author had to accept, lest no sample at all would be drawn. Since thi3 is considered an exploratory study, such limitations become less significant. 371 geographical and a socioeconomic criterion, some stratifi cation can be expected as a result of differences among those states. Therefore, the Brazilian sample has the virtue of some differentiation. The American sample is comprised of 137 indivi duals: 34 federal employees, 52 state employees, 50 local employees, and 1 unidentified. The federal employees are middle-level executives in a research and development organization dedicated to the creation and development of military weapons. American state employees work for the State of California, in Sacramento. They come from seven different agencies, as follows: (1) State Personnel Board, (2) Board of Equalization, (3) Department of Em ployment, (4) Department of Finance, (5) Department of Mental Hygiene, (6) Department of Public Works, and (7) Department of Natural Resources. Finally, the local employees included in the American sample work for the cities of Los Angeles and Inglewood and are mostly middle- level officials. As can be seen, the two samples are not comparable on several grounds. However, they have two characteris tics in common. The first is that all Americans and 372 Brazilians are public employees. The second similarity regards the occupational characteristics of the two groups and is very important for the following comparisons. Al most all American and Brazilian samples share an interest in academic training in public administration. In the case of Americans, the sample is comprised of trainees attending courses in public administration. Also, the majority of Brazilians were sampled at the Brazilian School of Public Administration, in Rio de Janeiro, and the School of Administration, in Bahia, where they were undergoing specialized training in administration. In regard to the personal background of the respondents, the two samples differed significantly from each other in all three variables under analysis. Brazilians were younger than Americans, a result which seems to reflect the age composition of the person nel in the two bureaucracies. The data on age were dis tributed as shown in Table 10. 373 TABLE 10 AGE Time Intervals (Years) Group 10-19 20-29 30-39 40-49 50-59 60-69 N Brazilians .01 .24 .47 .23 .05 .00 113 Americans .00 .05 .27 .40 .26 .02 136 X^ * 93.10 (categories 1 and 2, and 5 and 6 collapsed) P<^ .001 The results on age are compatible with those on supervision experience. Here, almost all Americans (93 per cent) have had some experience of supervision, in com parison to 57 per cent of Brazilians, as shown in Table 11. TABLE 11 SUPERVISION EXPERIENCE GrouD Yes No N Brazilians .57 .43 111 Americans .93 .07 137 X2 = 43.24 P C .001 374 Finally, among those with supervision experience, Americans also have had a longer experience than Brazil- 37 ians. The data on this variable are shown in Table 12. TABLE 12 LENGTH OF SUPERVISION EXPERIENCE Time Intervals (Years) Less than More than Groun 10 10-19 20-29 30 N Brazilians .79 .19 .02 .00 63 Americans .53 .38 .07 .02 127 = 12.15 (categories 2, 3, and 4 collapsed) p <1 .005 These differences between the two samples seem to be consistent with each other and reflect actual differ ences between the two bureaucracies. Indeed, this author would expect Brazilians public employees to be younger in average than Americans, and also less experienced in supervisory functions in general. ^In all tables of this chapter, the data are presented in the form of proportions. Data Processing and Statistical 375 Methodology Once the questionnaires were collected} the data were punched on IBM cards. For the majority of the items, the coding followed the same order in which the alterna tive responses to each item were presented in the instru ment. Thus, coding categories had to be developed for a few cases only, as can be seen in Appendix D. The measure of inferential statistics applied to test the differences between the two groups was chi square. Every item of the questionnaire, including those on background information, was subjected to that test of significance. Whenever marginal frequencies in the original distributions of the data were too low to provide satisfactory expected frequencies for the chi square test, adjacent categories were collapsed in order to meet this requirement. ^®See Chapter V, footnote 87. 376 Findings of the Study In order to facilitate its understanding, this section is structured in terms of the three temporal variables under analysis, namely: (1) perception of time as a scarce commodity; (2) concern for scheduling; and (3) personal versus organizational orientation in the alloca tion of work hours. A conclusion follows the analysis of the findings on the three variables. Perception of Time as a Scarce Commodity The first three items— 5, 6, and 7— dealing with this variable focus on timeliness or punctuality, both on business matters and on social affairs. It is submitted that one's definition of lateness, and reaction to it as well, is indicative of the extent to which one is con cerned with optimizing the use of his time. The data on item 5 are presented in Table 13, in the form of proportions. 377 TABLE 13 DEFINITION OF LATENESS Time Intervals (Minutes) GrouD 0 -09 10-19 20 or more N Brazilians .22 .57 .21 115 Americans .52 .44 .04 137 X2 = 30.57 The chi square test indicates that those differ ences are significant (p .001). Also, an examination of the direction of the relationship shows that Americans define lateness in more strict terms than Brazilians. Though not shown in Table 13 (for they were collapsed into the 20-29 interval in order to satisfy statistical re quirements) , two Brazilians considered an arrival as late only if it occurred sixty or more minutes after the expected time. The following item dealt with tolerance, rather than definition of lateness: How long would one wait for a latecomer? The differences between the groups are not significant (p .95). Also, though a higher proportion of Brazilians fell in the more optimizing end of the 378 continuum, the differences do not provide a clear indica tion of which group tends to be more tolerant about late- ness. The data on this item are distributed TABLE 14 TOLERANCE OF LATENESS in Table 14. Time Intervals ('Minutes) Group 0-19 20-29 30-39 40-49 More than 50 N Brazilians .31 .14 .38 .08 .09 115 Americans .28 .13 .42 .08 .09 137 X2 = .95 Item 7 inquired about the respondent’s emotional reaction to lateness on social matters, supposing that the latecomer were a friend. Here again the two groups do not differ significantly from each other (p .75), nor are there solid grounds for diagnosing the direction of the relationship. The data which pertain to this item are shown in Table 15. 379 TABLE 15 REACTION TO LATENESS (SOCIAL MATTERS) Grout) Angry Indifferent Pleased N Brazilians .17 .65 .18 115 Americans .16 . 64 .20 137 X2 = .51 On the basis of the three above items on timeli- ness, it is concluded that, although the two groups do not differ in their reactions about lateness, they do define it in different terms. Thus, both Brazilians and Americans would wait about the same time for a latecomer to an appointment on business matters, but Americans have much narrower limits for acceptable tardiness. This implies the existence of more sanctions and less deviance in temporal commitments, for individuals are expected to operate within shorter temporal ranges. Observe that item 7 deals not with organizational behavior, but with social matters; it also refers to espachic relationships, for the latecomer is a friend of the respondent, In this case, the groups reacted as it would be expected, i.e., in a 380 very similar way. To an extent, the results on the three items suggest that Brazilians tend to carry to the organi zational sphere the same temporal laxity of their social behaviors, whereas Americans tend to be more stringent about organizational time. Item 8 dealt with the length of the official work week for government employees in the two bureaucracies. It is submitted that the amount of time required from occupants of public jobs is indicative of how scarce time is perceived in a b u r e a u c r a c y . ^ A thirty-hour week and a forty-hour week, for instance, reveal different percep tions of the availability of time as a scarce commodity. As the data show, the differences are highly significant (p .001), Brazilian public employees being required a shorter work week than Americans (see Table 16). OQ Actually, this information might have been obtained directly from statutes and service orders in Brazil and the United States. However, the question was included in the instrument, among other reasons because this would facilitate the collection of the data on the matter in Brazil. 381 TABLE 16 PRESCRIBED WORKING LOAD Time Intervals (Hours') GrouD 20-29 30-39 Over 40 N Brazilians .04 .74 .22 103 Americans .00 .00 1.00 135 = 163.969 (categories 1 and 2 collapsed) The above results show that holding of a public position in the United States is a full-time job, which is 40 not the case in Brazil. Actually, four Brazilian respondents indicated that their official work week was between 20 and 29 hours. Observe that this item is not confined to perception of time as a scarce commodity. Indeed, it is also related to personal versus organiza tional orientation in the allocation of work hours, as will be discussed later. 40The current position classification plan for the Brazilian federal government (Law 3780, Collec^o de Leis do Brazil. July 12, 1960) provides for full-time regime of work in technical-scientific positions. The employee is left the option about working full-time, in which case he will have a salary increase of at least 75 per cent, and in some cases of 125 per cent. 382 The following two items— 9 and 10— dealt with sense of urgency or hastiness. Thus, time compulsion was considered as an indication of perception of time as a scarce commodity, a point which may also be linked to the notion of high achievement motivation.^ Item 9 focused on the respondents' felt sense of urgency or hastiness. The two groups differed significantly from each other in this regard (p ^ .001), Americans feeling usually more rushed than Brazilians. This is shown in Table 17. TABLE 17 RESPONDENTS' FELT SENSE OF URGENCY Group <-- More Less - ---K N Brazilians .17 .30 .37 .13 111 Americans .32 .48 .13 .04 130 X2 = 30.761 Item 10 inquired about the respondents' percep tion of governmental personnel's efforts to do as much as ^In his The Achieving Society (Princeton: Van Nostrand, 1961), p. 324, ff., David C. McClelland presents considerable evidence supporting the relationship between high n achievement and perception of time as a scarce commodity. 383 possible in tbeir available time. In other words, commit ment to a high work pace was taken as a form of time com pulsion, and, therefore, of time optimization. Here again, Brazilians and Americans differ significantly from each other (p .001), the latter appearing more fre quently at the optimizing end of the continuum. The data on this item are shown in Table 18. TABLE 18 GOVERNMENTAL PERSONNEL’ S WORK PACE Group --Faster Slower — — y N Brazilians .06 .10 .48 .32 107 Americans .12 .60 .19 .04 121 X2 = 90.240 The remaining items— 11, 12, and 13— referred to the respondents’ perception of the actual, not official, time inputs of governmental personnel. It seems unneces sary to develop any justification of the relevance of such a point to the perception of time as a scarce commodity in 384 Ao a bureaucracy. * In the first case, the respondents were asked to provide a rough guess of how many hours per week top professional people in the government are actually on the job. The data on this item (11) took the distribu tion shown in Table 19. TABLE 19 TIME INPUTS OF TOP GOVERNMENTAL PERSONNEL Time Intervals (Hours^ GrouD• 10-19 20-29 30-39 40-49 50-59 60-69 N Brazilians .31 .28 .15 .13 .10 .04 94 Americans .00 .01 .05 .30 .46 .19 133 X2 = 121.136 The above differences are statistically signifi cant (p .001) and, as far as the direction of the relationship is concerned, American top professionals work more hours than their Brazilian counterparts. Indeed, most Americans work between 50 and 59 hours a week, in contrast to only 10 per cent of Brazilians. The coraposi- ^These three items are related to personal versus organizational orientation in the allocation of work hours, also. However, this point is left to the appropri ate section, below. 385 tion of the American sample— particularly due to the samplees from the research and development organization— may have biased such results considerably, so that they represent a higher average than the one which holds for the bureaucracy as a whole. However, this would not affect the differences between the two groups. These results are critical for the process of administrative reform, since so much of it depends on the higher echelons of the bureaucracy. When the strategy for a given change relies on a chain of command whose occupants do not meet the time inputs required for such a change, the results are not likely to be satisfactory. The following two items— 12 and 13— probed pos sible differences between the governmental and the business subcultures, in their perceptions of time as a scarce commodity. This was done by way of comparing the organi zational time inputs of public and private personnel, at both higher and lower echelons. The rationale behind those items was that, since business organizations tend to develop an optimizing-organizational pattern of temporal orientations as a requisite of their own survival, a com parison of the relative time inputs of personnel in 386 government and in business might provide an indication of the extent to vhich time is perceived as a scarce good in the governmental subculture. The data on items 12 and 13 are shown in Tables 20 and 21. TABLE 20 RELATIVE TIME INPUTS BETWEEN PUBLIC AND PERSONNEL (HIGHER ECHELONS) PRIVATE Public Personnel Spend More Time Less Time Group on the Job on the Job . N Brazilians .01 .03 .25 .36 .35 114 Americans .07 .14 .59 .16 .04 128 X^ a 69.242 (categories 1 and 2 collapsed) p .001 TABLE 21 RELATIVE TIME INPUTS BETWEEN PUBLIC AND PRIVATE PERSONNEL (LOWER ECHELONS) Group Public Personnel Spend More Time on the Job Less Time on the Job -*■ N Brazilians .01 .00 .24 .41 .34 114 Americans .02 .06 .79 .11 .02 132 X2 = 98.76 (categories 1 and 2 collapsed) p^: .001 387 The results above reveal striking differences be tween the two groups. In regard to higher echelons, the data reinforce the conclusion arrived at in item 11 that the actual time inputs of Brazilian top governmental per sonnel are rather limited. As expected, the majority of Americans (59 per cent) sees no differences between the time inputs of public and private personnel at higher echelons. In other words, time is perceived as scarce in the governmental subculture as in the business milieu. This is not the perception of Brazilians, for whom top government personnel spend less time (36 per cent) or con siderably less time (35 per cent) than business personnel on the job. When one turns to the lower echelons, the same sort of relationship is found. Again, the bulk of Americans is concentrated on the third response category (79 per cent), and Brazilians on the fourth (41 per cent) and fifth (34 per cent). In both items 12 and 13, it is quite apparent that Brazilians tend to perceive governmental employees at both higher and lower echelons as spending less time on the job than their business counterparts, whereas Americans think 388 that the time inputs of government and business personnel are about the same, at any organizational level. There* fore, one may conclude that, if time is not perceived as a scarce commodity in Brazil, it is even less so in the governmental subculture. The implications of such a fact for a strategy of change which requires optimizing temporal orientations are self-evident. Conclusion On the basis of the nine items dealing with per ception of time as a scarce commodity— items 5 to 13— the null hypotheses about the perceptions of Brazilians and Americans must be rejected. Indeed, the two groups differed from each other in 7 out of 9 items. Also, in regard to the direction of the relationship, Brazilians fell in the less optimizing cells with a higher frequency than Americans in 7 out of 9 cases, a proportion which, submitted to the sign test, is statistically significant (p = ■ .02). Therefore, this temporal requirement of the instrumental raodel--perception of time as a scarce commod ity— is not a prevailing orientation of the client system. In other words, it is not a feature of the behavioral profile of the Brazilian bureaucracy. 389 Concern for Scheduling This variable was the subject of ten items— 14 to 23— of the questionnaire. They focused on different aspects of concern for scheduling. The first four items— 14, 15, 16, and 17— dealt with scheduling habits, reactions to lack of scheduling, and expectations or values about scheduling. In item 14, the respondents were asked how often they received people without any prior appointment. The answers on this item are shown in Table 22. TABLE 22 SCHEDULING HABIT (APPOINTMENTS) Group ^------More Less — > N Brazilians .31 .21 .28 .20 113 Americans .50 .46 .04 .00 137 X2 = 68.629 The two groups are significantly different from each other (p ^ .001), but the direction of the rela tionship is opposite to what would be expected: Americans receive people without any appointment more frequently than Brazilians. If considered more carefully, however, these results cannot be taken as indicative of a greater commitment of Brazilians than of Americans to scheduling habits. The Brazilian sample is comprised mainly of individuals whose positions do not involve receiving people, or setting appointments. Therefore, the Brazil ians who answered "occasionally" (28 per cent) or "not at all" (20 per cent) might well be saying that they usually do not receive people, the question of previous appoint ment being irrelevant here. On the other hand, the American sample included many individuals whose positions necessarily involve receiving higher authorities often without any prior appointment. This is the case, for instance, of department heads. When one turns to reactions to lack of scheduling, however, it is interesting that Brazilians tend to be less disturbed than Americans. Though the differences between the two groups were not significant (p .25) in item 15, more Americans (25 per cent) than Brazilians (19 per cent) fell in the "disturbed" category, as seen in Table 23. 391 TABLE 23 REACTION TO LACK OF SCHEDULING (APPOINTMENTS) Group Disturbed Indifferent Pleased N Brazilians .19 .71 .11 112 Americans .25 .68 .08 136 X2 = 1.99 Therefore, though Americans must receive people without prior appointment because of the nature of their positions, they nevertheless feel more disturbed than Brazilians about such a practice. This seems to indicate that the commitment to scheduling is greater among the former than among the latter. The following two items dealt with perceptions of, and expectations for, scheduling in regard to the respond ents' typical work day. Thus, in item 16, the respondents were asked whether they felt that their typical work day proceeded according to an orderly plan (item 16) and if they felt that it should proceed according to such a plan (item 17). As in the case of authority patterns, percep tions were to be considered as actual behaviors, and 392 expectations as values. The data on these two items are presented in Tables 24 and 25. TABLE 24 SCHEDULING HABIT (TYPICAL FORK DAY— PERCEPTIONS) Group More Less ,, ■,, v N \ ■ ' r Brazilians Americans .22 .00 .35 .42 .28 .46 .15 114 .12 137 X2 = 36.652 TABLE 25 SCHEDULING HABIT (TYPICAL FORK DAY— EXPECTATIONS) Group More Less v N Brazilians Americans .20 .08 .37 .49 41 .01 43 .00 .01 109 .00 136 X2 = 11.282 (categories 3, 4, and 5 collapsed) The differences above are significant in both items 16 (p .001) and 17 (p .005). In the former, the direction of the relationship is opposite to what would be expected, Brazilians appearing as working more 393 under an orderly plan than Americans. However, as in the case of appointment in the reception of people (item 15), these may well be positional, rather than cultural, dif ferences. Indeed, they may simply reveal a difference between the more programable positions of the Brazilian sample, as compared with the less programable ones of the American sample. It is indicative of such a point, for instance, that no American answered that his typical work day "always" proceeds according to an orderly plan, which was the case of 22 per cent of Brazilians. As to item 17, the fact that the two groups are not significantly different from each other from a statis tical standpoint is highly significant from the perspective of the theoretical rationale of this study. This is so because that item is supposed to reflect values about scheduling. Therefore, even if, on the basis of the nature of the positions which comprise the two samples, one disregards the fact that more Brazilians (20 per cent) than Americans (8 per cent) fell in the response category most indicative of a higher prizing of scheduling, it still remains a fact that the two groups are very similarly concentrated on the several response categories. When such 394 results are contrasted with the others having to do with scheduling, one may suggest that, though the groups dif fer as to actual behaviors, such differences are not apparent in the case of values. As may be recalled, the data on authority patterns led to similar conclusions. From the standpoint of strategy of change, this is doubly significant in regard to Brazil. First, because it implies the existence of potential forces which might facilitate the change process, even though such forces, in the form of values, have not been translated into actual behaviors. Second, because it supports the criticism that the instrumental model, by being cognitively, rather than valuationally, oriented, has failed to take strategic advantage of those change forces that exist in a latent state in the client system. The following item— 18— dealt with span of time perspective. It is submitted that concern for scheduling is directly related to a long-run time perspective, one which is long enough to involve futurity, and short enough to be realistic. Indeed, immediatism and scheduling seem contradictory almost by definition. On the other hand, 395 Asian futuristic orientations considering centuries ahead are equally inimical to scheduling. As pointed out by an American: In a sense, we are interested in that which we can schedule, in a time space where the data are cer tain enough to make synchronization a reasonable possibility. Thus, ten years ahead is at the end of our planning spectrum; we definitely are not like the Asiatics who are much more likely to perceive the future in terms of generations or centuries.43 Thus, in item 18, the respondents were asked about their time perspective in planning and thinking about their jub future. The data on this item are shown in Table 26. TABLE 26 RESPONDENTS' TIME PERSPECTIVE Group Never Think About Future Next Month Next Year Next Three Years Next Ten Years Beyond Next Ten Years N Brazilians .38 .29 .26 .06 .01 .00 110 Americans .02 .07 .26 .41 .22 .02 136 9 = 115.53 (categories 5 and 6 collapsed) P < .001 ^ S h e r w o o d , pp. cit.. p. 6. See, also: Hall, op. cit., p. 134. 396 These figures allow for several interesting con clusions about Brazilians and Americans' time perspective. First, 38 per cent of Brazilians said that they never think about the future, as contrasted to only 2 per cent of Americans. Second, the great majority of Americans (89 per cent) was concentrated in the third, fourth, and fifth categories, which cover a time span of ten years. The heaviest concentration is on the third category (41 per cent), which refers to the next three years. There fore, the contention that Americans tend to confine their temporal perspectives within schedulable limits is clearly supported by the data. Contrariwise, after the "never think about the future" category, the greatest concentra tion of Brazilians is on the first category (29 per cent), which refers to the next month. From the standpoint of a strategy for organiza tional change, a relevant point may be developed in rela tion to the short time perspective of Brazilians. It re fers to the extent to which this characteristic has been reinforced by the prevailing strategic pattern of the reforms. Organizational research has brought forth con siderable evidence of an association between an j 397 individual's short time perspective and his lack of control over his environment. Argyris has been prominent among those who have explored this problem. More than an association, he establishes a causal nexus between the two characteristics: As a result [of formal organization requirements] the individuals have little control over their working environment. Concommitantly, their time perspective is shortened because they do not control the informa tion necessary to predict their future.44 Insofar as such a general proposition can be applied to specific cases, the strategy that has guided the Brazilian administrative reforms is inadequate, to say the least. Indeed, since that strategy has been characterized by a centralization-and-control approach, it has increas ingly reduced the decisional autonomy of the field agencies even on minor matters. The reforms have not constituted a developmental experience for those agencies, in the sense of enabling them to control and decide on their own 44Chris Argyris, Personality and Organization (New York and Evanston: Harper & Row, Publishers, 1957), pp. 60-61. Italics in the original. This point is also developed in his more recent, Integrating the Individual and the Organization (New York: John Wiley & Sons, Inc., 1964), p. 44. These books report several empirical find ings on the relationship between short time perspective and one's lack of control of his environment. 398 matters. Instead of responsible initiative and willing ness to take decisions, the reforms have generated passive dependence. To the extent that such a reduction of the field agencies' control of their own environment stimu lates a short time perspective, the reforms have actually developed resistances which are functions of their strat egy. To conclude, if the short time perspective of Brazil ians is predominantly a cultural characteristic— an impedi ment— the strategy embodied in the instrumental model has reinforced it. On the other hand, if that characteristic is predominantly organizational and has at least in part resulted from the strategy itself— an incompatibility— the adequacy of such a strategy is open to even more severe criticisms.^ In both cases, the strategy would be foster ing orientations which are antagonistic to those orienta tions which the strategy requires for the change to succeed. The remaining five items— 19 to 23— dealing with concern for scheduling focused not on span, but on direc tion, of time perspective. The rationale behind these 1 items is that concern for scheduling implies a futuristic ^The distinction between impediment and incom patibility is rationalized in Chapter I, pp. 10-14. 399 orientation, rather than a time perspective directed to the past. Thus, item 19 dealt with direction of time perspec tive in governmental decision-making. The respondents were asked which considerations— past, present, or future— guided them the most, at the time of making a decision. The distributions of the answers are shown in Table 27. TABLE 27 DIRECTION OF TIME PERSPECTIVE IN GOVERNMENTAL DECISION-MAKING Direction N X or z P JErgsent -Official. EaULcy Brazilians 95 1.474 .742 2.822 < .005 Americans 123 1.805 .989 Past Precedents Brazilians 86 3.116 1.056 3.433 <. .008 Americans 115 3.565 .690 Future Imnlications Brazilians 93 2.710 1.028 1.215 = .2224 Americans 122 2.541 .989 Note: N = sample size; x « mean; cT = standard deviation; z = standard score; and p * probability. 400 As can be seen, the two groups differ signifi cantly from each other (p .005) in their attachment to present official policy as their decisional basis. 46 Brazilians ranked this alternative higher than Americans. In regard to past precedents, the differences were also significant (p .008) Brazilians again attributing a greater importance to this alternative than Americans. As to future implications as a decisional basis, however, though the differences were not significant (p ~ .2224), Americans ranked higher than Brazilians. All in all, the data on this item reveal that governmental decision-making in Brazil is less oriented toward the future, and more toward the past and present, than in the United States. From the standpoint of concern for scheduling, this implies that the decisional process in Brazil is less related to goals and deadlines than £n the United States. ^Since i corresponds to the most important, and 3 to the least important, alternative In the rank order for this item, a smaller mean indicates a higher rank order and a larger mean indicates a lower rank order. The meas ure of inferential statistics applied to this item was the test of the difference between two means. See, in this regard: Helen M. Walker and Joseph Lev, Statistical Inference (New York: Henry Holt & Co., 1953), p. 155. 401 Items 20 and 21 dealt with the respondents' per ceptions of, and expectations for, the temporal orienta tions of their governments, respectively. The data on these items are presented in Tables 28 and 29. TABLE 28 GOVERNMENTAL TEMPORAL ORIENTATIONS— PERCEPTIONS Group Past Present Future N Brazilians .09 .52 .38 107 Americans .01 .54 .45 134 X2 = 7.983 GOVERNMENTAL TABLE 29 TEMPORAL ORIENTATIONS---EXPECTATIONS Group Past Present Future N Brazilians .00 .38 .62 113 Americans .00 .34 . 66 134 X2 * .534 (categories 1 and 2 collapsed) The chi square test shows that the two groups are significantly different from each other in regard to per ceptions (p .025), but not in regard to expectations (p .50). Brazilians perceive their government as more oriented toward the past than is the case with Americans. However, such an orientation is unanimously rejected by both Brazilians and Americans. That is, expectations (values) of Brazilians are much like those of Americans, even though their governments behave in different ways. As with the findings on authority patterns, the conflict with the instrumental model is sharper in regard to behav iors, and less pronounced (in this case, actually non existent) in regard to values. As noted before, this is an important strategic element which has been overlooked, since the instrumental model is cognitively, rather than valuationally, oriented. Finally, instead of focusing on the government as a whole, items 22 and 23 referred to the respondents' per ceptions of, and expectations for, the temporal orienta tions of their jobs. The distributions of the answers on these items are presented in Tables 30 and 31. 403 TABLE 30 RESPONDENTS' JOBS' TEMPORAL ORIENTATIONS— PERCEPTIONS Group Past Present Future N Brazilians .08 .78 .14 113 Americans .02 .66 .32 127 X2 = 12.605 TABLE 31 RESPONDENTS' JOBS' TEMPORAL ORIENTATIONS— EXPECTATIONS Group Past Present Future N Brazilians .00 .65 .35 108 Americans .01 .33 .66 135 X2 = 24.214 The two groups differ from each other as to both perceptions (p .005) and expectations (p .001). However, in the latter case, the differences are in regard to the "present" and "future" categories, not in regard to "past." Americans are more future-oriented and Brazilians more present-oriented in both their perceptions and 404 expectations. Therefore, the former have a longer time perspective than the latter. It is important to note that as in items 20 and 21, the differences between the two groups on the "past" category are more pronounced in the case of perceptions than in the case of expectations. Thus, the values of Brazilians and Americans in regard to temporal orientations of their governments and of their jobs toward the past are the same, though the actual orientation of both government and public jobs in Brazil is more attached to the past than is the case in the United States, Again, this supports the contention that there are change-oriented values in the Brazilian bureau cracy which have not been strategically exploited for the benefit of the reforms. The anti-valuational bias of the instrumental model may explain why such forces have been overlooked. Conclusion On the basis of the ten items dealing with concern for scheduling— items 14 to 23— the null hypotheses about Brazilians and Americans must be rejected. Indeed, the two groups 4iffeted from each other in 8 of the 10 items. 405 In regard to the direction of the relationship, Brazilians fell in the less optimizing cells with a higher frequency than Americans in 8 out of 10 cases, a proportion which, submitted to the sign test, is statistically significant (p s .011). Therefore, this temporal requirement of the instrumental model— concern for scheduling— is not a pre vailing orientation of the client system. In other words, it is not characteristic of the behavioral profile of the Brazilian bureaucracy. Personal versus Organizational Orientation in the Allocation of Work Hours This variable was the subject of five items— 24 to 28— of the questionnaire. Before turning to them, 4 of the items (8, 11, 12, and 13) dealing with perception of time as a scarce commodity will be discussed. Though such items were, on the basis of a given rationale, among those on perception of time as a scarce commodity, they are also related to the variable under analysis here. As will be recalled, the responses on item 8^ showed that government employment in Brazil does not 4?see Table 16. 406 involve the employee's full-time commitment to his job. To the extent that a part-time job allows (sometimes even stimulates) other work relations outside the public ser vice, those responses may suggest a personal orientation in the allocation of work hours in Brazil. This is so because to the extent (1) that an individual holds two jobs, one in the government and the other in business; (2) that his total working hours are not enough to fulfill all his obligations in the two jobs; and (3) that the probabil ity of losing the public job are— as in Brazil— almost nonexistent, such an individual will tend to sacrifice his time allocations to the public job in favor of the private one. The responses on item ll48 are compatible with those on item 8. Brazilian top governmental personnel work fewer hours per week than their American counterparts. Also, in the comparison between public and private top personnel (item 12) ,49 Brazilians perceive the former as spending less time or considerably less time than the 48See Table 19. 49See Table 20. 407 latter on the job. Similar results were found on item 13,^0 which dealt with the same matter, but focused on lower echelons. On the other hand, Americans generally perceive public and private personnel as spending about the same time on the job, both at higher and lower echel echelons. The results above are reinforced by the findings on items 25 to 28, which were specifically concerned with orientation in the allocation of work hours. Item 25 dealt with job-unrelated allocation of time on the job. The respondents were asked to estimate the percentage of their work time spent with people in activities of a social nature. The assumption about this question was that the less the time spent on job-unrelated matters, the stronger the organizational orientation in time allocation. The data on this item are shown in Table 32. 50see Table 21. 408 TABLE 32 JOB-UNRELATED ALLOCATION OF TIME ON THE JOB ReSDonse Categories (Percentages) Group Less than 10 10-25 26-50 More than 50 N Brazilians .68 .31 .01 .00 112 Americans .85 .15 .00 .00 134 = 10.810 (categories 2, 3, and 4 collapsed) The differences between the two groups are signi ficant (p ^ .001), and a higher proportion of Brazilians fell in the less optimizing cell than Americans. One Brazilian answered that he spends between 26 per cent and 50 per cent of his working time on job-unrelated activities. The following four items— 25 to 28— dealt with the question of government officials holding dual employment. It is reasonable to assume that, as long as an employee works only for one employer, the probabilities of his using working hours for personal purposes are smaller than would be the case if he were a part-time employee who had other jobs. This is so because of the ultimate scarcity 409 of time for both the individual and the organization. Dual employment increases the conflict which is inherent in such scarcity. The individual may then have to decide as to whether to sacrifice his time allocations to organi zation A or those to organization B. As noted before, in the case of Brazil, a number of reasons suggest that, in case of conflict between two jobs, the employee will tend to sacrifice the one which is not likely to involve penalties— the governmental job. Items 25 and 26 dealt with dual employment in government, at higher echelons. Asked about the common* ness of dual employment at those levels, the respondents provided the following answers (item 25), as shown in Table 33. TABLE 33 DUAL EMPLOYMENT IN GOVERNMENT (HIGHER ECHELONS) Group — More Common Less Common # ^ N Brazilians .52 .34 .09 .05 .00 114 Americans .00 .05 .03 .30 .63 134 X2 = 194.555 410 The responses leave no doubt that the notion of full-time employment does not prevail in the Brazilian public service at all. The differences between the two groups are highly significant (p <^.001). These results become more critical in view of the responses to item 26. The respondents who stated that it was either very common or fairly common for top professional people to hold dual employment were asked to estimate the percentage of time that such people spend on governmental business. The responses on this item are shown in Table 34. TABLE 34 TIME INPUTS AFFECTED BY DUAL EMPLOYMENT IN GOVERNMENT (HIGHER ECHELONS) Time Spent on Government Business (Percentages) Group Less than 24 25-49 50-74 More than 75 N Brazilians .07 .32 .42 .19 90 Americans .07 .07 .07 .79 28 = 36.36 (categories 1 and 2 collapsed) Again the two groups differ significantly from each other (p .001) and, in regard to the direction of the relationship, most Brazilians (42 per cent) see top 411 governmental personnel allocating no more than 74 per cent of their work time to government business. Also, for a considerable proportion of Brazilians (32 per cent), top personnel with dual employment allocate less than 50 per cent of their work hours to the government. An overwhelm ing majority (79 per cent) of the twenty-eight Americans who answered this question sees dual employment affecting no more than 25 per cent of the time of top governmental personnel. It seems unnecessary to expand on the significance of such results for administrative reform, particularly when that reform depends upon the prevalence of organiza tional, rather than personal, criteria for the allocation of time in the client system. The fact that items 25 and 26 focused on top personnel, who are usually responsible for any reform, makes this point a critical one. Finally, items 27 and 28 also dealt with dual employment, but at the lower levels. The distributions of responses on these items are very similar to those in the case of higher echelons, as can be seen in Tables 35 and 36. 412 TABLE 35 DUAL EMPLOYMENT IN GOVERNMENT (LOWER ECHELONS) Group .. More Common Less Common — ^ N Brazilians .29 .26 .24 .15 .06 110 Americans .01 .26 .00 .58 .15 132 X2 = 102.918 TABLE 36 TIME INPUTS AFFECTED BY DUAL EMPLOYMENT IN GOVERNMENT (LOWER ECHELONS) Time Spent on Government Business (Percentages) Group Less than 24 25-49 50-74 More than 75 N Brazilians .08 .10 .48 .33 60 Americans .07 .00 .13 .80 43 - 25.276 (categories 1 and 2 collapsed) The groups differ significantly from each other in both item 27 (p .001) and item 28 (p .001), As in the case of higher echelons, dual employment is more common in Brazil than in the United States, and the con sequences on the employee's allocation of time to govern- 413 mental affairs is perceived as much greater in Brazil. Considering that the control of time allocation in the United States is incomparably greater than in Brazil, such results were expected to occur. Finally, it is worth noticing that, according to the data, dual employment is more frequent at higher, than at lower echelons, in Brazil, whereas in the United States the reverse is true. Though the characteristics of the samples do not allow any gen eralization on this point, it deserves mention. Conclusion On the basis of the five items dealing with per sonal versus organizational orientation in time allocation, the null hypotheses about Americans and Brazilians must be rejected. Indeed, the two groups differed significantly from each other in all those items. Brazilians fell in the less optimizing cells with a higher frequency than Americans in all items also (p <^,001). Therefore, this temporal prerequisite of the instrumental model— organi zational orientation in time allocation— is not a preva lent characteristic of the behavioral profile of the Brazilian bureaucracy. 414 Summary and Conclusions This chapter investigated the extent to which the temporal requirements of the instrumental model of change strategy are fulfilled by the Brazilian bureaucracy. Those requirements were subsumed in the concept of "optimizing-organizational temporal orientations." Three indicators of an optimizing-organizational pattern of temporal orientations were developed, namely, perception of time as a scarce commodity, concern for scheduling, and personal versus organizational orientation in the alloca tion of work hours. These indicators were the bases for the three hypotheses which governed the study, as enumer ated below: 1. There is no significant difference between public administrators in Brazil and the United States in regard to their perception of time as a scarce commodity. 2. There is no significant difference between public administrators in Brazil and the United States in regard to their concern for scheduling. 3. There is no significant difference between public administrators in Brazil and the United States in 415 regard to their use of work hours for organizational, as against personal, purposes. Data on twenty-eight items about these hypotheses were collected and, on the basis of their analysis and interpretation, the three hypotheses were rejected. In deed, Brazilians and Americans differ significantly from each other as to their time orientations. Differences are sharper in the case of perceptions or actual behaviors, than in expectations or values about time-related behav iors. In general, Brazilians hold less optimizing and more personal orientations toward time, as contrasted with Americans. In relation to the theoretical framework of this dissertation, the findings of the present study indicate that the prevailing time patterns in the Brazilian bureau cracy are antagonistic to those which are required by the strategy of change that has guided the attempts to reform that bureaucracy. Thus, the client system does not ful fill the prerequisites of the change strategy. It is worth noticing, though, that in several items dealing with values, rather than behaviors, Americans 416 and Brazilians were either slightly different from each other, or no differences were found. This suggests the existence of values which would be supportive of bureau cratic reform and which would be accounted for by valua- tionally-oriented strategies. It also suggests that less optimizing and personnally-oriented time patterns are not as much expected or desired as they are practiced. Thus, to the extent that such expectations or values are dis seminated in the client system, its degree of changeful ness will be positively affected. Since the instrumental model is cognitively oriented, these change-supportive values have not been exploited from a strategic stand point . CHAPTER VII CONCLUSIONS This chapter is organized into three parts* The first integrates what has been said up to this point. The second applies the findings of the study to an interpreta tion of the Brazilian experience of administrative reform. And the last is an attempt to suggest guidelines for the future based on the lessons of the past. Summary This is a study in planned administrative change, from the perspective of comparative public administration. It focuses on the adequacy of change strategies, given the behavioral characteristics of a client system. The major theoretical proposition guiding the study is that the greater the compatibility between the requirements of a strategy of change as to the behavioral patterns of the client system, on the one hand, and the patterns which 417 418 actually shape the behavioral profile of that system, on the other, the greater the strategic adequacy— and, conse quently, the probabilities for success— of an intended change. The client system under analysis is the Brazilian federal bureaucracy, which since the early thirties has undergone numerous attempts at planned administrative change. The specific aims of this study are (1) to analyze and evaluate these attempts from the perspective of the adequacy of the strategies employed and, on the basis of such an analysis (2) to suggest guidelines for the devising of strategies for future reform efforts in the Brazilian bureaucracy. To achieve such ends, it was necessary to identify and describe (1) the prevailing strategic patterns of administrative reform in Brazil and (2) the actual behavioral profile of the bureaucracy, so that change strategy and client system might then be related to each other. As noted before, the analysis adopted a comparative approach. The United States was chosen as the referent for comparison with Brazil, a choice justified by the fact that American administrative thought and practice have pervaded the Brazilian reform 419 experience. It seems only logical to select for compari son the country whence the rationale for, and the processes of— and this implies saying, the values of— administrative reform in Brazil have been sought. 'The identification and description of the change strategies which have prevailed in the Brazilian reform experience involved the devising of a model of change strategy which would hypothetically embrace the predomin ant strategic characteristics of the reforms. In the formulation of such a model— which was called the instru mental model of change strategy— it was assumed that any strategy is analytically reducible to three components: (1) valuational, (2) substantive, and (3) operational. The valuational component refers to assumptions, beliefs, preferences, and attitudes which underlie a strategy. This was called strategy ideology and considered as the most important of the components, because it conditions the other two. The substantive component refers to the actual content of a strategy, and was called strategy content. Finally, the operational component refers to the characteristics of the change agent actions in the imple mentation of a strategy, and was called strategy behavior. 420 The instrumental model consists of (1) a strategy ideology which requires functional position-orientedness in executive behavior and in values about executive behavior, and an optimizing-organizational pattern of temporal orientations; (2) a strategy content character ized by a wholistic, rather than gradualistic, change scope, by a prescriptive, rather than suggestive, language style, and by a cognitively, rather than valuationally, oriented obstacle-mapping process; and *3) by a strategy behavior focused on formal, rather than informal, elements. The analysis of the behavioral profile of the Brazilian bureaucracy centered on two of the most critical traits of the profile of any organization, namely, authority and time patterns. Brazilian public adminis trators were compared with Americans, in regard to these traits. The comparison was based on empirical data collected by means of two questionnaires, one dealing with executive behavior and the other with tenporal orienta tions. The total sample for the two studies consisted of 589 individuals, of which 212 were Brazilians and 377 were Americans. 421 Seven hypotheses were then developed as the basis for relating the change strategy to the behavioral profile of the client system. The two major hypotheses submitted that (1) administrative reforms in Brazil have been patterned after an instrumental, rather than a noninstru mental, model of change strategy, and that (2) the behav ioral patterns required by the strategy ideology of the instrumental model are incongruent with those which characterize the behavioral profile of the bureaucracy. The five minor hypotheses centered on authority and time patterns as traits of the behavioral profiles of the Brazilian and the American bureaucracies. These hypotheses were all stated in the null form, as follows: There is no significant difference between Brazilian and American public administrators in regard to their (1) perceptions of, and (2) expectations for, executive behavior, (3) perceptions of time as a scarce commodity, (4) concern for scheduling, and (5) personal versus organizational allocation of work hours. The first major hypothesis was supported, on the basis of an analysis of the Brazilian experience of administrative reform since 1930. Indeed, the analysis 422 revealed that the prevailing strategies employed in that experience have conformed to the instrumental model, as defined above. In order to ascertain whether the requirements of the strategy ideology of the instrumental model are ful filled by the Brazilian bureaucracy— and thereby test the second major hypothesis— the five minor hypotheses had to be tested. These were all rejected. Indeed, the data showed that Brazilians and Americans differ significantly from each other on every one of the minor hypotheses. Specifically, it was found (1) that Brazilians perceive and (2) expect dysfunctional executive behaviors, both of a positional and of a nonpositional type, and (3) do not perceive time as as scarce a commodity, (4) have a lesser concern for scheduling, and (5) have a personal, rather than organizational, orientation in the allocation of work hours, as contrasted with Americans. These findings then supported the second major hypothesis that the behavioral patterns required by the strategy ideology of the instru mental model are incongruent with those which characterize the behavioral profile of the bureaucracy. 423 In short, the study shows that the reform experi ence in Brazil has depended upon a prerequisite-based change strategy and that such prerequisites are not characteristic of the behavioral profile of the Brazilian bureaucracy. An Interpretation of the Past The most difficult obstacles to administrative reform in Brazil have been of a political nature. Indeed, it is during the legislative phase that the majority of reform attempts has been halted. As suggested before, some of the very characteristics of the instrumental model which have hampered the implementation of the reforms may well have accentuated their political hurdles. For example, a wholistic approach may demand administrative capabilities beyond those actually available in the bur eaucracy and represent unattainable political compromises as well. The focus of this study, however, is not on the strategy of political breakthrough, but on the strategy of administrative implementation, once reform proposals have been approved by law. The basic questions, then, are: 424 Why is it that, even when freed of any legislative obstruction (as during the dictatorship), the implementa tion of the reforms was still far from satisfactory? What can be learned in regard to administrative planned change from the retrogression that, since the redemocratization of the country in 1946, has plagued whatever accomplish ments the reforms had made? One possible answer to these questions is that the behavioral prerequisites on which the reform efforts depended do not prevail in the client system. In other words, the behavioral requirements of the instrumental model of change strategy are not characteristic of the behavioral profile of the Brazilian bureaucracy. This is the explanation that the present study suggests for the failures of administrative reform in Brazil. The argument, however, does not stop here. An interminable series of questions might ensue. Some of them have been answered earlier. Others are worth more thought. It seems helpful— at least in regard to the Brazil ian case— to distinguish between the strategy and the teleology of the reform movement. Indeed, one might con- 425 ceive the general ideology of the reforms as consisting of two components, the strategic and the teleological. This should not imply any dichotomy between means and ends, or similar philosophical points. With this caution in mind, one may proceed with distinct critiques for the strategy and the teleology of the reforms. From the latter perspective, it is hardly arguable that, at least in principle, the reforms were in the "right" direction. They embodied a commitment to widening the horizons of administrative rationality in the civil service. They aimed at replacing ascription with achieve ment, particularism with universalism, corruption with morality, waste with economy, and malfunctioning with efficiency in the administration of the public affairs. From a strategic perspective, however, the reforms were inadequate for at least three reasons. First, they resorted to what has been called else where "an essentially prerequisite-based strategy of change," the prerequisites of which are not prevalent characteristics of the client system. In other words, the strategy patterned after the instrumental model required given behavioral patterns to prevail in the 426 client system if the intended change were to succeed. Since this was not the case, the probabilities for success of the change attempt were reduced, which opens the way to criticisms as to the adequacy of such a strategy. To be sure, the fact that a strategy is based on prerequisites does not necessarily imply its inadequacy. The crucial point is whether such prerequisites prevail in the client system. Second, the strategic inadequacy of the Brazilian administrative reforms occurs because the strategy did not provide for devices to germinate or disseminate the values conducive to the behaviors required. This is a much more complex and relevant point, so far as change strategy is concerned. On the basis of this point, one might suggest that the existence of a conflict between the behavioral requirements of the strategy and the behavioral realities of the client system was not crucial for the fate of the change attempt. The real shortcoming, from the strategic standpoint, was the failure to buttress the reform process with devices by which the behavioral patterns of the ■ client system would be made more compatible with the strategy employed. Actually, in some cases the strategy 427 seemed to evoke, or at least reinforce, behaviors antagon istic to those which the strategy itself required. For example, the centralization bias of the reforms, by reduc ing the field agencies' control of their own environment, may have contributed to the strengthening of the short time perspective of the Brazilian bureaucrat. Since the instrumental model requires a long-time perspective, it may be suggested that, by reinforcing a short-time per spective, the strategy has operated against its own behavioral requirements. * • The third and last point about the strategic inadequacy of the reforms is the most important, for this analysis. It concerns the extent to which the strategy has affected the teleology of the reforms in a feedback fashion. As noted before, the teleology of the reforms has been "correct,’ ' at least in principle. The "at least in principle" clause is very important for, depending on the dynamics of each specific case, that teleology may well turn out to be inadequate or dysfunctional. It is suggested that this is exactly what has happened in *See Chapter VI, pp. 396-98. 428 Brazil. The development of this point requires a brief theoretical digression. Not a few scholars have criticized the analytical partiality of bureaucratic theory, in the sense that only its functional aspects were highlighted. Today, it has become a commonplace in administrative theory that bureau cracy generates functions and dysfunctions. To go fur ther, every bureaucratic element is potentially both functional and dysfunctional, for the simple reason that bureaucracy has contradictory functional requirements.* Admitting that this applies to bureaucratic reform, it then becomes necessary to analyze which aspects— function al or dysfunctional— of the Brazilian reform have tended to be more accepted, and how the strategy employed has influenced such an acceptance. It is submitted that the instrumental model, the goals of which can hardly be criticized, has nevertheless resulted in significant dysfunctions for the bureaucracy. This is so because (1) the behavioral profile of the bur eaucracy is characterized by dysfunctional patterns of ^For an elaboration of this point, see the section "Functionality and Dysfunctionality of Organizational Behavior," Chapter IV, pp. 253-68. 429 executive behavior and of temporal orientations, and (2) the anti-valuational bias of the instrumental model has precluded any systematic effort at inculcating values which are more compatible with the attempted changes. The bureaucracy has tended to incorporate only those aspects of the reforms which are congruent to it— the dysfunction al aspects, therefore--and to reject those which are in congruent— the functional ones. Many examples illustrate the above point, but the most impressive are found in the personnel reforms. The same law that regulated the merit system in the federal service^ determined its demise as well. The Federal Constitution of 1934. which first made entry to the public service subject to competitive examinations, referred to public jobs as "positions" (cargos). However, because of political pressures and the fear of unemployment, and also because the Executive himself needed a greater flexibility in the elimination and creation of public jobs, the law created a new juridical figure, the "function" (fun^o). "Positions" were created by the Legislative, while ^Law 284, Coll.. October 28, 1936. 430 "functions" might be created by Executive decree. In consequence, occupants of "functions" (called "extra- numeraries" or "supernumeraries") were not subjected to the same set of rules applicable to occupants of "positions" (called "functionaries").^ In short, "extranumeraries" were not subjected to entry examinations by the constitu tion. An entirely different terminology was then invented for the new category, so as to create a parallel, inde pendent group of employees. In line with the legalistic tradition, juridical technicalities lent the indispensable endorsement of the law to the new state of affairs. The following comparison is illustrative:^ ^This distinction came to an end in 1960, with Law 3780, Coll.. July 12, 1960. The solutions given to this problem were themselves highly formalistic. This, however, is not a pertinent point here. ^Translation as literal as possible. 431 Functionary Occupies a position Position is created by law Entry by examination Is appointed Is promoted Receives remuneration (remuneracao) Positions form career (carreira) Careers form cadre (guadrg) Bxtranumerarv Exerts a function Function is created by executive decree Entry by proof of capability Is admitted Is not promotable but gets salary increase Receives salary (palario) Functions form functional series (serie funcional- ) Functional series form table (tabel§) The artificiality of the distinction is punctuated in a well known joke in Brazilian administrative circles: the functionary dies, the extranumerary "gives up the ghost." The relevant point is that, created in 1936 under a complementary character, the figure of the extranumer ary increased incessantly, so that by 1960, for about 100,000 functionaries there were about 200,000 extranumer- aries. The attempt to give the Executive a greater flex ibility in the elimination and creation of jobs had a 432 dysfunctional, unintended consequence. Less bound to merit system rules, extranumeraries enjoyed more chances of ascending the civil service ladder, better salary scales, and other advantages. Although higher in status, many functionaries gave up their "positions" in favor of "functions," thus turning into extranumeraries. The above example of the absorption of the dys functional, and the rejection of the functional, aspects of the teleology of the reforms is replicated in other areas. In budget, the establishment of an a posteriori control in order to facilitate financial operations has given place to many corrupt practices. Position classifi cation reforms have failed to subordinate pay policies to a rational distribution of duties and responsibilities; yet these reforms have been used as instruments to give legal sanction to old privileges and open the way to new ones. In sum, because the strategy has not been directed at building support for the reforms by modifying the dysfunctional patterns which prevail in the behavioral profile of the bureaucracy, there has been a tendency for the client system to incorporate the dysfunctional aspects 433 of Che reforms. For a related reason, the functional aspects of the reforms have tended to be rejected or blocked. Thus, not only the strategy but (as a consequence of it) the teleology of the reforms as well is inadequate, so far as changing the Brazilian bureaucracy is concerned. Actually, one might suggest that, in a sense, the reforms have not elicited innovation, but conservation. In other words, instead of actually changing the client system, they have contributed a great deal to the stabilization of the status q u o , by providing opportunities for the rein forcement of patterns that already prevail in the behav ioral profile of the bureaucracy. A Proposal for the Future The interpretation of the past provides a basis for devising strategic guidelines for future attempts at administrative reform. Before passing from one to the other, it is worth reflecting a while on the present. The following remarks on what might be called biases of the present are not based on the data. It is submitted, however, that such remarks add to the explana tion of the past and introduce discussion about the future. 434 Since the instrumental model has proved to be of a limited adequacy to the reform cause, one might wonder why the model has survived and still guides reform attempts. There are many answers to this point, at dif ferent theoretical levels. A simple but convincing reason is that the group of change agents responsible for the reforms has remained basically the same. At a higher level of abstraction, however, one may hypothesize that the instrumental model finds support in two biases, or perspectives, towards the development of public adminis tration as an intellectual discipline and as a governmen tal function. Such perspectives are separable but inter related. It is suggested that they still have consider able receptivity in the academic circles of administration in Brazil, though new orientations seem to be emerging. The first refers to a unilinear perspective of administrative development. The most common stereotype of this perspective is that, both in theory and practice, administration has moved from scientific management to human relationism to structuralism. Such moves are per ceived by some as being incremental and by others as antithetical to each other. The relevant point for the 435 present discussion is that this perspective implies a bias about change which has strategic implications of the high est importance. Indeed, those who subscribe to this view tend to believe, for instance, that human relations is only feasible after a system has "lived through" scienti fic management. After all, "it was so in the United States." In this case, since Brazil "has barely entered the human relations stage," behavioralism is an approach for the future, not for the present. In other words, only after the bureaucracy achieves a satisfactory degree of formal systematization is it possible to focus on behav ioral changes. The reverse process— that of using behav ioral strategies to achieve some degree of bureaucratiza tion— is hardly accepted by those who share this perspec tive. They have difficulties understanding that only by employing a Bennis can the Brazilian bureaucracy maxi mize the benefits from Gulick's "Posdcorb," The second perspective resembles the first one, but in a broader context. There has been some tendency among those who take a sociological approach to adminis tration to consider it as a dependent variable in the social process. For them, administrative changes are but reflections of social transformations* particularly those of an infra-structural nature. In this case, public administration is not at all significant as an agent of change. The implications of this perspective for change strategy are self-evident. Perhaps the best example of this approach to the role of public administration is the rationalization of the spoils system on the basis of ecological imperatives. Though such a rationalization is quite pertinent for some societies at given stages of their development, the fact is that, In the case of Brazil, it may be more academically favored than empirically sup ported. Indeed, it may well be that the Brazilian federal bureaucracy has overcome the stage where spoils are a necessary valve for alleviating pressures of political clientelism. The social system may not need that ' ’ remedy" any longer, though it keeps on "ingesting" it as a systemic vice. The political function of spoils has lost consider able importance with the industrialization of the country, and the political system may be able to survive without the spoils valve. If so, any insistence on rationalizing the need for spoils is dysfunctional to change. In sum, those who adopt this perspective fail to recognize that 437 public administration may be as much an elicitor, as it is an object, of social change. Their strategy is to wait for an appropriate time when changes would occur naturally. If any strategy is to be supported at all, this would more likely be of an instrumental sort, though confined to a few areas. As noted before, in spite of the considerable acceptance of the two perspectives just discussed, new outlooks are emerging, particularly in the teaching of administration in Brazil. These reflect a behavioral perspective, sometimes of an exaggerated human relations, others more integrative of the formal and informal ele ments. To the extent that such perspectives, particularly of the latter variety, gain more academic acceptance and come to influence the reform process, the hopes for administrative reform will be strengthened. It may then be that the change strategy to be adopted will follow the general guidelines suggested below. The analysis cf the past provides a sound basis for proposing strategies of change for future attempts at administrative reform in Brazil. That analysis offered an interpretation of the failures of reform efforts based on 438 (1) the incompatibility between the behavioral require ments of the strategy and the behavioral profile of the client system, (2) the nonexistence of devices strategic ally applied to modify the profile of the bureaucracy as necessary, and (3) the ultimate dysfunctionalism of the teleology of the reforms, in that their positive aspects have not found a response in the bureaucracy, whereas their negative ones have. Thus, the inadequacies of the strategy have reduced the possibilities for success of the change attempts. It is suggested that the reform process as a whole will increase its strategic adequacy— and, consequently, its probabilities for success— if it is oriented by the following guidelines: 1. Gradualistic orientation. The past has shown that the demands imposed by a wholistic change scope are beyond the existing administrative capabilities of the bureaucracy. During dictatorial periods, the nonexistence of legislative barriers has been an especially compelling inducement to global changes. However, even if entirely free to plan and proclaim a wholistic reform, the bureau cracy is not prepared to meet the attendant administrative 439 demands* The result has been formalism, with real losses for the idea of reform. As noted before, a discredited merit system may be worse than a limited but effective one. The degree of comprehensiveness of such gradualism must occur on a case-by-case basis. The main point is to keep the reform scope within realistic boundaries. 2. Selectivity. The constituent parts of any bureaucracy are unevenly change prone. Some are highly stable and virtually untouchable, others are adaptive enough to dispense with any outside help, and still others may be under varying degrees of potential disequilibrium and might experience rapid development if activated by the change agent. Past reforms in Brazil have not taken account of such differences. Rather, their wholistic orientation has promoted across-the-board undertakings, with a consequent misallocation of resources to areas which either (1) are irrelevant for the reform process as a whole, at a given stage, or (2) are so resistant to change as to be "unproductive investments" at a given mom ent, or (3) both. By adopting a selective orientation, change efforts would be concentrated on more change-prone areas, or on those which were critical .to .change., no 440 matter their degree of changefulness. Such an opportun istic approach would likely result in a better allocation of the scarce resources of the bureaucracy, the assumption being that once change is unleashed in a given part of the system, that part becomes a source of change inputs for other parts. 3. Continuity. The experience of the past has been characterized by intermittency, rather than contin uity, in change efforts. Reforms have usually been under taken under conditions of crisis, when the dysfunctions of fcke status quo have reached unbearable points. Much of their ineffectiveness has resulted from this intermittency— everything is started all over again, from time to time, by succeeding reform groups. There is a need for approaching change as a continuous, rather than intermittent or crisis, task. Continuity is directly related to gradualism. A gradualistic approach represents, in fact, a piece-meal change strategy. Its basic advantage is its political feasibility and, from an administrative standpoint, its realistic demands upon the client system. Its main weak ness is the slower rate of change it impels. However, the 441 latter point does not have to be necessarily so, as long as the reform process is continuous. A related aspect of the past is that reforms have reflected a formulation of change as a once-for-all, fully integrated process, rather than a cumulative one. This seems to be another symptom of the short time perspective that prevails in the bureaucracy. Here, again, the moti- vation-outruns-understanding approach of the strategy may have reinforced a behavioral predisposition— short time perspective— which is antagonistic to the requirements of the strategy itself. 4. Valuational orientation. One of the most dis tinctive characteristics of the instrumental model is its obstacle-mapping process geared to cognitive, rather than valuational, barriers to change. Hence, even the training programs related to the reforms have reflected a ration alistic bias and stressed a "how-to-do" approach. The data on both traits of the behavioral profile of the bur eaucracy revealed that the values which shape that profile are not conducive to the patterns of behavior that the instrumental strategy requires. It then becomes necessary to shift the focus of those training programs so as to 442 link a valuational orientation to the cognitive one which has prevailed. Another reason for this shift— also on the basis of the empirical analysis— is that the conflict between the instrumental model and the client system is sharp in regard to actual behaviors, but much less appar ent in terms of values. This is indicative of the exist ence of potential forces of change which could be exploit ed. Change-supportive values which already exist in the bureaucracy would then be identified and disseminated, so as to strengthen rejection of status q u o values. 5. Internalizing orientation. Experience has demonstrated that whatever were the achievements of the reforms during dictatorial periods in Brazil, those achievements entered a progressive stage of dilution, as soon as the controlling power was reduced by the political redemocratization of the country. This is so because influence processes based upon compliance require perman ent control by the change agent to maintain the intended change. Once control is removed, the system goes back in large measure to its previous state. Therefore, in a true sense, change has not occurred. This is what explains the relative success of the reform movement during the 443 dictatorship, and its debacle after the re-constitution of a democratic regime. The instrumental model did not provide for the internalization of the values which would give continuity to the reform work. In view of the current political situation in Brazil, the quest for internalization, rather than com pliance, takes on a renewed importance for the reform process. In a way, there has been a return to the past, since the change agents can resort again to compliance processes in order to effect change. The point, then, is how to reconcile the immediate need for compliance with the need for internalizing the values of the new regime, so as to assure the endurance of the achievements of the April Revolution. From the standpoint of a strategy of reform, this may constitute the most critical dilemma of the Brazilian bureaucracy of today. There are indications, for instance, that an ongoing reform in the Ministry of Finance has achieved excellent results— tax revenues have increased substantially, efficient control systems have made tax collection easier and safer for both the taxpayer and the public employee, and so on. This reform has adopted a gradualistic change scope. It has also reflected a selectivistic orientation, as the term is defined above, not in the sense of concentrating on a more changeful area of the client system, but on one where change was an imperative condition for the political stability of the new regime and for the performance of the bureaucracy as well* Nevertheless, as had to occur under the circum stances, the change agents have enjoyed a compliance power which sooner or later will cease. The test of the strate gic adequacy of the reform will occur when that power is removed. It is suggested that, by simply recognizing the need for internalization, change agents will become more sensitive to the possibilities of increasing the endurance of today's achievements by instilling their corresponding values into the client system. 6. Broadened leadership. This guideline involves an anti-elitist orientation to the strategy of a given change. The experience of the past was marked by an insulative pattern in the leadership of the reform process. A handful of technicians, an administrative elite, under took the task of reform. Thus, even though a network of subunits in charge of the reforms was formally created— the ministerial Denartamentos de Administrac&b and the 445 personnel of such units belonged to the ministries, the necessary leadership cooptation did not occur. Therefore, the organization structured to carry out the reforms never did constitute a system in the sense of having central and peripheral units which were homogeneous as to the values, purposes, professional qualifications, and loyalty orien tations. Actually, the change group tended to be a solitary star within the bureaucracy. From the perspective of change strategy, this insulation of the reform leadership bears heavily upon the development of resistances against the intended changes and in the coalescence of such resistances into a united front against the small reforming elite. It is submitted that an expansion of the leadership in charge of the reforms, especially if accompanied by systematic efforts to inculcate in it an adaptive change orientation is likely to result in a greater support for the change process. The above guidelines are by no means specific or comprehensive enough to constitute a model of change strategy for administrative reform in Brazil. This would require a degree of prescriptiveness for which this 446 exploratory study does not provide sufficient support. However, if compared to the prevailing strategic charac teristics of the Brazilian reforms, these guidelines represent a shift from an instrumental-rational to a behavioral concept of change strategy. It is suggested that this shift will increase the strategic adequacy of future reform efforts and, therefore, their possibilities of success. The Brazilian bureaucracy has developed many rigidities. These ossified areas need to become more flexible, more adaptive, and less bureaucratized. In other points, however, the problem is reverse, the need being one of bureaucratization. In any case, the need is to build a reform strategy from the realities of Brazilian behavior. BIBLIOGRAPHY BIBLIOGRAPHY Books Almond, Gabriel, and Coleman, James (eds.). The Politics of Developing Areas. Princeton, N.J.: Princeton University Press, 1960. Alves, Fdbio de Carvalho. As Atividades de Organizaqfifo na Administraqao Pdblica Brasileira (Rio de Janeiro: Imprensa Nacional, 1950. Argyris, Chris. Integrating the Individual and the Organization. New York: John Wiley & Sons, Inc., 1964. \ ________ . Personality and Organization. New York and Evanston: Harper & Row, Publishers, 1957. Baleeiro, Aliomar. Uma Introduqao a Ciencia das Financas. 2 vols. Rio de Janeiro: Editdra Forense, 1958. Barnard, Chester. The Functions of the Executive. Cambridge: Harvard University Press, 1938. Bennis, Warren G., Benne, Kenneth D., and Chin, Robert (eds.). The Planning of Change. New York: Holt, Rinehart & Winston, Inc., 1961. Berger, Morroe. Bureaucracy and Society in Modern Egypt: A Study of the Higher Civil Service. Princeton: Princeton University Press, 1957. Blau, Peter M., and Scott, V, Richard. Formal Organiza tions. San Francisco: Chandler Publishing Co., 1962. 448 449 Boliens, John C. Appointed Executive Local Governments: The Californian Experience. Los Angeles: The Haynes Foundation, 1952. Braybrooke, David, and Lindblom, Charles E. A Strategy of Decision. New York: The Free Press of Glencoe, 1963. Caplow, Theodore. Principles of Organization. New York: Harcourt, Brace & ForId, Inc., 1964. Comte, Auguste. System of Positive Polity. Vol. ZV. London: Longmans, Green & Co., 1877. de Roure, Agenor. 0 Oryamento. Rio de Janeiro: Pimenta de Melo, 1925. Erikson, Erik. Childhood and Society. New York: Norton, 1950. Etzioni, Amitai and Eva (eds.). Social Change— Sources. Patterns, and Consequences. New York: Basic Books, Inc., Publishers, 1964. Follet, Mary Parker. Dynamic Administration: The Col lected Forks of Mary Parker Follet. Edited by H. C. Metcalf and L. Urwick. New York: Harper & Bros., 1940. Gama e Silva, Jose Saldanha, and Pereira, Daisy Florie Passarinho. "0 Sistema de Orgaios Auxiliares e de Staff na Administra^ao Brasileira," Seminario Intemacional SBbre Organizacao. Direcffo e Funcionamento dos Servicos Auxiliares e OrgKos de Estado-Maior (Staff). Sdrie C, Tema 1. Rio de Janeiro: Funda^gfo Getulio Vargas and UNESCO, 1952. Gerth, H. H., and Mills C. Fright. From Max Feber. Trans lation with an introduction by H. H. Gerth and C. Fright Mills, New York: Oxford University Press, 1958. 450 Gould, Julius, and Kolb, William K. (eds.)* A Dictionary of the Social Sciences. New York: The Free Press of Glencoe, 1964. Gross, Neal, Mason, V. S., and McEachern, A. W. Explora tions in Role Analysis. New York: John Wiley & Sons, 1956. Hagen, Everett E. On the Theory of Social Change: How Economic Growth Begins. Homewood, 111.: Dorsey Press, 1962. Hall, Edward T. The Silent Language. Greenwich, Conn.: Fawcett Publications, 19S5. Hirschman, Albert 0. Journeys Toward Progress. New York: The Twentieth Century Fund, Inc., 1963. ________ . The Strategy of Economic Development. New Haven and London: Yale University Press, 1958. Homans, George C. The Human Group. New York: Harcourt, Brace & Co., 1950. Hugo, Victor, Les Miserables. Translated by Charles E. Wilbour. New York: The Book League of America, 1943. Jones, Garth N. (ed.). Planned Organizational Change— A Set of Working Documents. Los Angeles: Center for Research in Public Organization, School of Public Administration, University of Southern California, 1964. Kardiner, Abram. The Individual and His Society. New York: Columbia University Press, 1939. La Piere, Richard T. Social Change. New York: McGraw- Hill Book Co., 1965. Lasswell, Harold, and Kaplan, Abraham. Power and Society. New Haven: Yale University Press, 1950. 451 Lawrence, Paul R. The Changing of Organizational Behavior Patterns— A Case Study of Decentralization. Boston; Harvard University, Graduate School of Business Administration, Division of Research, 1958. Leavitt, Harold J. Managerial Psychology. 2d ed. Chicago and London: The University of Chicago Press, 1964. Lewin, Rurt. A Dynamic Theory of Personality. New York; McGraw-Hill Book Co., 1935. Likert, Rensis. New Patterns of Management. New York: McGraw-Hill Book Co., Inc., 1961. Linton, Ralph. The Cultural Background of Personality. New York: D. Appleton-Century Co., 1945. Lippitt, Ronald, Fatson, Jeanne, and Festley, Bruce. The Dynamics of Planned Change. New York: Harcourt, Brace & World, Inc., 1958. Macridis, Roy. The Study of Comparative Government. New York: Random House, Inc., 1955. Marx, Fritz Morstein. The Administrative State. Chicago: The University of Chicago Press, 1961. Marx, Karl. A Contribution to the Critique of Political Economy. Chicago: Charles H. Kerr, 1904. Mayo, Elton. The Human Problems of an Industrial Civili zation. New York: Viking, 1931. McClelland, David C. The Achieving Society. Princeton, N.J.: D. Van Nostrand, 1961. Merton, Robert K. Social Theory and Social Structure. Rev. ed. Glencoe, 111.: The Free Press of Glencoe, 1963. 452 Merton, Robert K., et al. (eds.). Reader in Bureaucracy. Glencoe, 111.: The Free Press of Glencoe, 1952. Mills, C. Wright. The Sociological Imagination. New York: Oxford University Press, 1959. Moojen, Guilherme. Orcamento Publico. Rio de Janeiro: Ediydes Financeiras S.A., 1959. Moore, Wilbert E. Man. Time and Society. New York: John Wiley & Sons, 1963. . Social Change. Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1963. Nascimento, Kleber. C^assificayao de Cargos no Brasil. Vol. XIII. Sdrie de Administracdb Conmarada. Rio de Janeiro: Instituto Brasileiro de Ciencias Administrativas, 1962. Newcomb, Theodore M. Personality and Social Change. New York: Dryden Press, 1943. Parsons, Talcott. Essays in Sociological Theory. Rev. ed. Glencoe, 111.: The Free Press of Glencoe, 1954. ________ (ed.). The Theory of Social and Economic Organization. Translated by A. M. Henderson and Talcott Parsons. Glencoe, 111.: Free Press and Falcon's Wing Press, 1947. Pfiffner, John, and Sherwood, Frank P. Administrative Organization. Englewood Cliffs, N.J.: Prentice- Hall, Inc., 1960. Piaget, Jean. The Moral Judgement of the Child. New York: Harcourt, Brace & Co., 1932. Presthus, Robert. The Organizational Society. New York: Alfred A. Knopf, 193?. 453 Pye, Lucien F. Politics. Personality, and Nation-Building: Burma1s Search for Identity. New Haven: Yale University Press, 1962. Riggs, Fred V. Administration in Developing Countries: The Theory of Prismatic Society. Boston: Houghton Mifflin Co., 1964. ________ . The Ecology of Public Administration. New Delhi: Asia Publishing House, 1961. Roethlisberger, F. J., and Dickson, F. J. Management and the Worker. Cambridge, Mass.: Harvard University, 1939. Rostow, W. W. The Stages of Economic Growth: A Non communist Manifest. Cambridge: Cambridge Univer sity Press, 1960. Selltiz, Claire, et al. Research Methods in Social Rela tions. New York: Holt, Rinehart & Finston, 1962. Sherif, Muzafer. The Psychology of Social Norms. New York: Harper & Bros., 1936. Sorokin, Pitirim A. Sociocultural Causality. Space. Time. New York: Russel & Russel, Inc., 1954. Spencer, Herbert. Sociology. Vol. I. New York: Appleton f i e Co., 1892. Spengler, Oswald. The Decline of the Vest. New York: Alfred A. Knopf, Inc., 1926. Stouffer, S. A., et al. The American Soldier. Princeton, N.J.: Princeton University Press, 1949. Toennies, Ferdinand. Community and Society; Gemeinschaft and Gesellschaft. Translated and edited by Charles P. Loomis. East Lansing, Mich.: The Michigan State University Press, 1957. 454 Toynbee, Arnold J. A Study of History. 6 vols. Abridged by D. C. Somervell. New York: Oxford University Press, 1946. Van Riper, Paul P. History of the United States Civil Service. Evanston, 111.: Row, Peterson &Co., 1958. Vieira da Cunha, Mario Wagner. 0 Sistema Administrativo Braslieiro. Rio de Janeiro: MinistArio da EducacjAb e Culture, 1963. Wahrlich, Beatriz M. S. Administrate de Pessoal— Principios e Tecnicas. Rio de Janeiro: Funda?ao Getdlio Vargas, 1964. Waldo, Dwight. The Administrative State: A Study of the Political Theory of American Public Administration. New York: The Ronald Press Co., 1948. Willoughby, V. F. Principles of Public Administration. Baltimore: John Hopkins Press, 1927. Periodicals. ArticlesT and Newspaper Apter, David E. "System, Process, and the Politics of Economic Development," Industrialization and Society. Edited by Bert F. Hoselitz and Wilbert E. Moore. Mouton: UNESCO, 1953. "Beltrlo na URGS— Reforma Administrativa federal nao incluioMinist£rio da Previd£ncia," Correio do Povo. September 25, 1965. Bennis, Warren G, "A New Role for the Behavioral Sciences: Effecting Organizational Change," Administrative Science Quarterly. VIII (1963), 125-65. 455 Bennis, Warren G. "A Typology of Change Processes," The Planning of Change. Edited by Warren G. Bennis, Kenneth D. Benne, and Robert Chin. New York: Holt, Rinehart, & Winston, Inc., 1961. Berlinck, E. L. "A Reforma dos Servi^s de Material," Revista do Service Ptiblico. Ill (July, 1940), 127- 30~ 9 Bierstedt, R. K. "The Problem of Authority," Freedom and Control in Modem Society. Edited by Morroe Berger, Theodore Abel, and C. H. Page. New York: Van Nostrand, 1954. Bogardus, Emory S. "The Concept of Gradualism," Sociology and Social Research. XLIV, No. 2 (November- December, 1959), 119-24. Chin, Robert. "The Utility of System Models and Develop mental Models for Practitioners," The Planning of Change. Edited by Warren Bennis, Kenneth D. Benne, and Robert Chin. New York: Holt, Rine hart & Winston, Inc., 1951. Dobyns, Henry F. "The Strategic Importance of Enlighten ment and Skill for Power," The American Behavioral Scientist. VIII, No. 7 (March, 1965), 24. Frank, Lawrence K, "Time Perspectives," Journal of Social Philosophy. IV, No, 4 (July, 1939), 294. Friedrich, Carl J. "Some Observations on Weber's Analysis of Bureaucracy," Reader in Bureaucracy. Edited by Robert K. Merton et al. Glencoe, 111.: The Free Press of Glencoe, 1952. Goodwin, H. F. "Work Simplification," Factory Management and Maintenance. July, 1958, p. 72. Gouldner, Alvin W. "Explorations in Applied Social Science," Social Problems. Ill, No. 3 (January, 1956), 169-81. 456 Gouldner, Alvin W. "Organizational Analysis," Sociology Today— Problems and Prospects. Edited by Robert K. Merton, Leonard Broom, and Leonard S. Cottrell, Jr. New York: Basic Books, Inc., 1959. Hallowell, Irving. "Temporal Orientation in Western Civilization and in a Pre-Literary Society," The American Anthropologist. XXXIX (October, 1937), 647. Heady, Ferrel. "Recent Literature on Comparative Public Administration," Administrative Science Quarterly. V (June, 1960), 134-54. Holmberg, Allan R. "Changing Community Attitudes and Values in Peru," Social Change in Latin America Today: Its Implications for United States Policy. Edited by Richard N. Adams et al. New York: Harper & Bros., I960. "The Changing Values and Institutions of Vicos in the Context of National Development," The American Behavioral Scientist. VIII, No. 7 (March, 1965), 3-8. Jones, Garth N. "Strategies and Tactics of Planned Organizational Change," Human Organization. XXIV, No. 3 (Fall, 1965), 199. and Aslam, Niaz. "Planned Organizational Change: A Classification Scheme for Analysis," Planned Organizational Change: A Set of Working Documents. Edited by Garth N. Jones. Los Angeles: Center for Research in Public Organiza tion, School of Public Administration, University of Southern California, 1964. "Strategies and Tactics of Planned Organiza tional Change: A Scheme of Working Concepts," Philippine Journal of Public Administration. VII, No. 4 (October, 1963), 275-85. 457 Kelman, Herbert C. “Processes of Opinion Change," The Public Opinion Quarterly. XXV, No. 1 (Spring, 1961), 57-81. McEachem, Alexander V7., and Palmer, Ted. “Effectiveness of Different Treatment Processes," View of Author ity: Probationers and Probation Officers. Report No. 1. Edited by A. V. McEachem. Los Angeles: Youth Studies Center, 1961. Mead, Margaret. "The Implications of Culture Change for Personality Development," American Journal of Ortho-Psychiatry. XVII (1947), 633-46. Moreira, Oscar Victorino. "Sistema de Material no Servi<;o Publico Federal," Revista do Service Publico. II (June, 1951), 36-42. Merton, Robert K. "Bureaucratic Structure and Personal ity," Social Theory and Social Structure. Edited by Robert K. Merton, Rev. ed. Glencoe, 111.; The Free Press of Glencoe, 1963. ________ . "The Unanticipated Consequences of Purposive Social Action," American Sociological Review. 1936, 894-904. Presthus, Robert V. "Behavior and Bureaucracy in Many Cultures," Public Administration Review. XIX, No. 1 (Winter, 1959), p. 34. Redfield, Robert. "The Folk Society," The American Journal of Sociology. LII (January, 1947), 293-308. Riggs, Fred W. "Bureaucrats and Political Development: A Paradoxical View," Bureaucracy and Political Development. Edited by Joseph La Palombara. Princeton, N.J.: Princeton University press, 1963. ________ . "Trends in the Comparative Study of Public Administration," International Review of Admin istrative Sciences. XXVIII, No. 1 (1962), 9-15. 458 Sanders, Irwin T. "Approaches to Social Change," Educa tion for Social Work. 1960 Proceedings, Eighth Annual Program Meeting, Council on Social Work Education, New York, New York. Address delivered in Oklahoma City, Oklahoma, January 20, 1960. Cited in Warren G. Bennis, Kenneth D. Benne, and Robert Chin. The Planning of Change. New York: Holt, Rinehart & Winston, 1961. Sherwood, Frank P. "The View from the Outside," Public Administration Review. V, No. 4 (December, 1963), 247-52. Shull, Fremont A., and Miller, Delbert C. "Help the Man in the Middle," Nations Business. LI, No. 2 (February, 1963), 36-37. Siegel, Gilbert B. "Administration, Values, and the Merit System in Brazil," Public Personnel Review. XXIV (July, 1963), 163. Sjoberg, Gideon. "Contradictory Functional Requirements and Social Systems," The Journal of Conflict Resolution. IV, No. 2 (June, 1960), 198-208. ________ . "Folks and 'Feudal' Societies," The American Journal of Sociology. LVIII (November, 1952), 231-39. Waldo, Dwight. "Review of the book Authority.". Edited by Carl J. Friedrich. Cambridge: Harvard University Press, 1958. In Administrative Science Quarterly. Ill, No. 2 (September, 1958), 268. Weber, Max. "Bureaucracy," From Max Weber: Essays in Sociology. Edited by H. H. Gerth and C. Wright Mills. New York: Oxford University Press, 1959. 459 Unpublished Materials Bennis, barren G. "Theory and Method in Applying Behav ioral Science to Planned Organizational Change." Unpublished paper, Alfred P. Sloan School of Management, Massachusetts Institute of Technology, n.d. (Mimeographed.) Hall, Jay, O'Leary, Vincent, and Williams, Martha. "The Decision-Making Grid: A Model of Decision-Making Styles." Unpublished paper of the National Parole Institutes, Washington, D.C., 1964. Khan Mohammed Huraayun. "Authority Behavior: An Empirical Approach to Comparative Public Administration." Unpublished Doctoral Dissertation, School of Public Administration, University of Southern California, 1965. McEachem, Alexander V., and Taylor, Edward M. "Inducing Change with Social Research— The Experimental Introduction of Research Findings within the Profession of Probation." Unpublished draft of a research proposal, Youth Studies Center, University of Southern California, May, 1965. Ministro Extraordinario para a Reforma Administrativa. "Anteprojeto de Lei e Exposi^efo de Mbtivos— Reorganiza o Sistema de Material de Servico Publico Civil, e Da^ Outras Providencias." Un published material, Rio de Janeiro, July, 1963. (Mimeographed.) ________ . "A Reforma Administrativa Federal de 1963." Speech before the House of Representatives, April 17, 1953, by the Extraordinary Ministry for Administrative Reform. (Mimeographed.) Niaz, Aslam. "Strategies of Planned Organizational Change." Unpublished Doctoral Dissertation, University of Southern California, 1963. 460 Peabody, Robert Lee. "Authority in Organizations: A Comparative Study." Unpublished Ph.D. disserta tion, Department of Political Science, Stanford University, i960. Pownall, G. A. "An Analysis of the Role of the Parole Supervision Officer." Unpublished Ph.D. disserta tion, Department of Sociology, University of Illinois, 1963. Santos, Hugo dos, and Farout, James. "Estudo das Atividades de Organizap&o e Metodos no Servi90 Publico Federal." Unpublished paper of the Organization and Methods Service of DASP, Rio de Janeiro, March, 1960. (Mimeographed.) Sherwood, Frank P. "Comparative Authority Orientations in the United States and Brazil: A Preliminary Probe." Unpublished paper delivered at a Work shop Conference for Trainers on New Research Find ings in the Behavioral Sciences held at the University of Southern California, Civic Center Campus, Los Angeles, April 30, 1964. ________. "The Budgeting Process in the Brazilian Federal Government." Unpublished paper written in Brazil, Getulio Vargas Foundation, May 29, 1963. ________ . "The Clock and the Specialized R and D Society," unpublished paper, 1965. (Dittoed.) Siegel, Gilbert B. "The Strategy of Public Administra tion Reform: The Case of Brazil." Unpublished paper, University of Southern California, Los Angeles, n.d. ________. "The Vicissitudes of Governmental Reform in Brazil: A Study of the DASP." Unpublished Doctoral dissertation, University of Pittsburgh, 1964. ________, and Nascimento, Kleber. "Formalism in Brazilian Administrative Reform: The Example of Position Classification," unpublished paper, June, 1965. 461 Public Documents Brazil. Decree 562, Coll.. December 31, 1935. Brazil. Decree 9941, Coll.. May 27, 1942. Brazil. Decree 11101, Coll.. December 11, 1942. Brazil. Decree 19512, Coll.. December 20, 1930. Brazil. Decree 19587, Coll., January 14, 1931. Brazil. Decree 20489, Coll.. January 24, 1946. Brazil. Decree 39855, Coll., August 24, 1956. Brazil. Decree 51705, Coll., February 14, 1963. Brazil. Decree-law 579, Coll.. July 30, 1938. Brazil. Decree-law 1184, Coll.. April 1, 1941. Brazil. Decree-law 1713, Coll.. October 10, 1939. Brazil. Decree-law 1909, Coll.. December 26, 1939 Brazil. Decree-law 2206, Coll.. May 20, 1940. Brazil. Decree-law 3838, Coll.. November 19, 1941 Brazil. Decree-law 5715, Coll.. July 31, 1943. Brazil. Decree-law 5984, Coll.. November 10, 1943 Brazil. Decree-law 8823A, Coll.. December 7, 1945 Brazil. Decree-law 9503, Diario Oficial. July 23, Brazil. Decree-law 3569, Diario Oficial. August Z \ 1946. , 1941. 462 Brazil. Federal Constitution of 1934. Arts. 169 and 170. Brazil. Federal Constitution of 1937. Art. 12. Brazil. Federal Constitution of 1946. Art. 186. Brazil. Imperial Constitution of 1824. Art. 172. Brazil. House of Representatives, Bill No. 4.844 (1954). Brazil. House of Representatives, Bill No. 3563 (1953). Brazil. House of Representatives, Bill No. 212 (1914). Brazil. House of Representatives, Bill No. 77 (1911). Brazil. House of Representatives, Bill No. 320 (1907). Brazil. Law 23, Coll.. October 30, 1891. Brazil. Law 284. Coll.. October 28. 1936. Brazil. Law 1584. Coll.. March 27. 1952. Brazil. Law .711, Coll.. October 28, 1952. Brazil. Law 3780, Coll., July 12, 1960. Brazil. Law 3807. Coll.. August 26. 1960, CEPA. _A Raforma Administrative Brasileira. Vol. I: Reoreanizayfe da Presid6ncia da Republica. Rio de Janeiro: Departamento de Imprensa Nacional, 1960. DASP. A Divisao de Oreapizac'Sb e Coordenacao e Suas Atividades eto 1942. 7Rio de Janeiro: Servi^o de DocumentacSb do DASP. 1943. > * DASP. Exoosicab de Motivos. No. 670, June 11, 1951. League of California Cities. Municipal Administrative Assistants. Berkeley: League of California Cities, 1951. Ministro Exfcraordinario para a Reforma Administrativa, Plano de Execuc3o da Reforma. Rio de Janeiro: Graftea Editora Livro, 1963. . Reforma Administrativa Federal de 1963 — Anteoroleto de Lei Organica do Sistema Adminis- trativo Federal. Vol. I: Exposicao de Motives. Brasilia: 1963. Presidente da Republica. Mensaeem Apresentada ao Congresso Nacional na Abertura da Terceira Sessao “ i a a i a “ £ — da Decima-Terceira Legislatura oelo Presidente da Reoublica V?ashinei^^J.uis Pereira de Sousa. Rio de Janeixo>-^j^^e^sa Nacional, 1929. ________ . M£nsas^"rTr 398 , de 6 de a gos to de 1954. do Poder Executivo. . . APPENDIX A AUTHORITY PATTERNS QUESTIONNAIRE AUTHORITY PATTERNS QUESTIONNAIRE PART I BACKGROUND INFORMATION Please fill in the following: 1. Your age ______________________________ 2. Marital status ________________________ 3. Number of children . _ 4. Father’s occupation_____________________________ 5. Your occupation _______________________________ 6. Number of years in this occupation _________ 7. Number of years of education ____________________ 8. What degrees do you hold? _______________________ 9. What was your major field of study? ____________ 10. Do you have any specialized training or education? 11. What is your religion? _____________________ 465 466 PART II INSTRUCTIONS Position-Oriented Behavior The following questions represent hypothetical situations involving a superior A and a subordinate B in a government organization. You are requested to indicate 1. What actually happens in each situation. 2. What you think should happen in each situation. 1. What actually happens. Your response should be based on your general observations of superior officer behavior, including your own. Clearly, the same behavior need not necessarily be exhibited in similar situations at all times, nor will all superior officers behave the same way in each situation. You are therefore asked to list the behavior you consider most likely in your own cultural environment. Since there is no measure of good or bad involved, please do not hesitate to mark the response that is most accurate in your view. It is not the purpose of this study to show that one type of authority behavior is better than another. 467 2. What should happen. In the second column, you are asked to give your view as to how A should behave. You will notice that whereas the first column requires you to report a fact, this second column requires you to state an opinion. Here again there is no test of goodness or badness involved so please do not hesitate to express your views frankly. One note of caution. In most individual situa tions between a superior and a subordinate there are extraneous factors that influence attitudes. For example, they may both have gone to the same school, or they are family friends, or they belong to the same official cadre. Please try to focus on the general "superior-subordinate" relationship rather than on individual cases in which regular behavior patterns will naturally be modified by extraneous factors. Also, while answering the questions you ought to try and think of subordinates no lower than three levels of rank below the superiors. Complete anonymity is assured. Please do not put your name or designation on the questionnaire. Your coop eration is deeply appreciated. 468 Each hypothetical situation is followed by some alternative behavioral responses. Choose the alternative which most closely corresponds to actual behavior and write the number assigned to that alternative on the line below the column "What actually happens." Next, choose the response which you think should be the one in each case and write the number of that alternative on the line below the column "What should happen." Status-Protectiveness 469 What Actually Happens 1, B, a subordinate, is generally re garded as more competent than A, his superior, and A knows this. ____ 1. A will find this an uncomfortable situation and will try to get rid of B by transfer or other means. 2. A might occasionally accept such a situation, but will not be very happy about it. 3. A might occasionally accept such a situation and even enjoy having such a competent subordinate. 4. A will welcome such a situation and consider himself fortunate to have such a competent subor dinate. He will not even think of making a comparison with himself. 2. B, a subordinate is rather difficult to get along with and has a tendency to question the orders of A, his superior. _____ 1. A will regard him as a nuisance to the organization and try to get rid of him. 2. A will regard him as a bad sub ordinate and exercise strict control over him. What Should Happen 470 What Actually Happens 3. A will not concern himself too much with B's habit of question ing his directives but will treat each objection on its merits, 4. A will welcome B's habit as a sign of independent thinking and con sider it an asset to the organi zation. 3. A issues a directive and B, the sub ordinate, thinks it is wrong. After a while A himself is not too sure, but all his other subordinates have received the directive and it is known throughout the office that B wants it changed. 1. A will try to maintain the dis cipline of the office and see that his directive is carried out. 2. A will weigh the advantages of maintaining discipline against the disadvantages of carrying out an incorrect directive and decide accordingly. 3. A will not be too concerned with the question of discipline and, if he is convinced, will not hesitate to accept B's suggestion. What Should Happen VThat What Actually Should Happens Happen A, the superior, is conducting an in spection of the work of his subordi nate B. He finds that he does not really understand it, although he is supposed to be familiar with it. 1. A will proceed with the inspec tion as though he understood and will try to give no indication of his inability. 2. A will put off the inspection on some excuse and divert attention to some other matter pertaining to B. 3. A will put off the inspection and tell B that he will first familiar ize himself with the work and then carry out the inspection. 4. A will ask B to explain those items which he cannot understand. B, a subordinate, possesses a special talent which is also a requisite of A's job, but A lacks it. 1. A would encourage B to develop this talent further in the interests of the organization and his own career. 2. A would utilize B's talents for the purposes of the organization with out being keen on giving too much personal credit to B. 472 What Actually Happens 3. A would attempt to play down B's talent because it shows him, the superior, up In a poor light. 6. B. the subordinate addresses A, his superior, by his first name. A would _____ 1. Find this objectionable. 2. Find this objectionable if other subordinates are present. 3. Not object, but not really like it. 4. Welcome the informality. 7. A wishes B to do a piece of official work. A would 1. Ask B to do it; e.g., MB, would you please do such and such?" 2. Tell B to do it; e.g., "B, do such and such." 8. A, the superior, walks into the office and finds his subordinate B at his desk. 1. A would expect B to stand up in all cases as a sign of deference. 2. A would expect B to stand up only in formal situations, e.g., an official meeting. What Should Happen 473 What Actually Happens 3. A would expect B to stand up, but only as a matter of social custom, just as he (A) might do if B walked in. 4. A would not expect B to stand. 9. Which of the following attitudes would you expect from A, who is presiding at a meeting of his own subordinates? _____ 1. He will tty and extract as much relevant information from the group as will enable him to arrive at the best decision. 2. He will try to share leadership with responsibility and will talce the risk of making all the decisions on the basis of his judgment. 3. He, as the head, will assume full responsibility and will take the risk of making all the decisions on the basis of his judgment. 10. A is by rank the senior-most official in a conference setting and is pre siding. Representatives of special ist departments are also present. During the conference _____ 1. A would allow the leadership of the discussion to pass to each official representative of an agency as and when a matter per taining to that agency arises. What Should Happen 474 What Actually Happens 2. A would identify the most compe tent man in the field under dis cussion even if he is not the official senior representative of the agency concerned and try to let him lead the discussion. 3. As chairman of the conference and the senior-most official present, A would see to it that the leader ship did not leave his hands at any point in the discussion. 11. Supposing A has to delegate some work in his organization. He would normally _ 1. Delegate it to the man next below him in the hierarchy. 2. Delegate it to the subordinate most expert in that field Irrespective of his position in the hierarchy. 3. Delegate it to a subordinate whom he learnt to trust the most be cause of a personal relationship. 12. C sends a complaint about his superior B to the head of the organization A, without first bringing it to B's notice. _ _ 1. A would examine the complaint, conduct independent investiga tions, and take action accord ingly. Vthat Should Happen 475 What Actually Happens 2. A would refer the complaint to B for his comments. 3. A would return it to C and in struct him to bring the com plaint to B's notice first. 4. A would penalize C for not bring ing it to B's notice first. 13. Supposing a member of the public out side the organization makes an appli cation to A directly for a certain service, e.g., a license. If A possesses all the facts but knows that normally such matters come to him through B, will he _____ 1. Take action independently as in all cases. 2. Send it on to B so that it comes back to A through proper channels. 3. Take independent action only if the applicant is an especially influ ential figure. The following is a simple organization chart. Please look at it and then answer ques tions 14 and 15 below. A T---------------- 1 B B rr It t—r 1 i cccc cccc What Should Happen 476 What Actually Happens 14.________________________________________________ 1. Normally speaking, A will treat persons of rank B with more courtesy than persons with rank C. 2. Normally speaking, A will treat all his subordinates the same way. 3. Depending on their competence, A will treat subordinates with pro portionate courtesy no matter what their position in the hierarchy. 4. A’s attitude towards his subor dinates will be conditioned more by factors like the social status of the subordinate, his education, his age, his membership of the same official cadre, et cetera, than by his position in the hierarchy. 15. If a person of the rank of B disagrees with the proposal of a person of the rank C, _____ 1. A is more likely to support B. 2. Since the level of specialization is higher at rank C, A will sup port C. 3. A will exercise his own best judge ment and decide according to the merits of the case without bother ing about hierarchical rank. What Should Happen 477 What Actually Happens 4. A will select certain persons in the organization in whom he has personal confidence and tend more to rely on them in most matters. The following questions are addressed to you directly, and it would be appreciated if you spoke from your per sonal experience only. 16. Mark the most valid statement. _____ 1. tfy subordinates obey my authority by virtue of the fact that I am their superior officer. 2. My subordinates obey my authority because they consider me more competent at the job than they. 3. My subordinates obey my authority because I have good relations with them and they respect me as a person. 17. Mark the most valid statement._____________ 1. By letting them know that if they cannot do a particular task, they can come to me know ing that I will be able to help them with it. What Should Happen 2. By letting them know that if they flout my authority, they will get into trouble. 478 What Actually Happens 3. By demonstrating, whenever time permits, how to do the task X have assigned to them. 18. I obey the orders of my superior because _____ 1. He is my superior. 2. He knows more about the job than I do. 3. I respect him as a person. 19. 1 am most likely to comply with the orders of my superior if 1. X know that he can do the job well himself and, therefore, X cannot deceive him. 2. X feel that if X do not, X am likely to get into trouble. 3. He and X are on good personal relations. Extension of the Scope of the Hierarchical Relationship 20. A, the superior, and B, the subordi nate are invited to the same party. At the party, 1. A would expect B not to try join the same conversational group as he and to keep at a distance. What Should fimaa 479 What Actually Happens 2. A would expect B to join his group, but to show him the same deference as he does in the office. 3. A would expect B to show him some deference, but not carry the official relationship totally to the social sphere. 4. A would expect B to treat him like any other social acquaintance and to forget that they have an official relationship. 21. Would A, the superior, expect the wife of his subordinate B to show defer ence in social relationships to his wife ? _____ 1. Yes. 2. No. 3. Indifferent, 22. From your experience, would you expect A, the superior, to ask B, his subordinate, to do private chores for him?_______________________ _____ 1. Regularly. 2. Quite often. 3. Very seldom. 4. Never. What Should Happen 480 What Actually Happens (Please remember to exclude any special personal relationships that may exist between A and B.) 23. B, a subordinate, who has private sources of income, owns a much bigger and more expensive motor car than A, his superior. . 1. A would tend to resent this. 2. A would not concern himself with this. 3. A might take notice of it, but not mind to any great extent. 24. B, a subordinate, who has private sources of income, moves in social circles that are rated higher than those in which A, his superior, moves. _____ 1. A would tend to resent this. 2. A would not concern himself with this. What Should Happen 3. A would tend to treat B with more courtesy and respect because of this. APPENDIX B AUTHORITY PATTERNS— CODING CATEGORIES CODING CATEGORIES The data from the questionnaires were punched on two separate cards. 1. Card No. 1 contained background information on the respondents, as obtained from Part I of the question naire . 2. Card No. 2 contained the responses to the twenty-four questions designed to test for position- orientedness, as obtained from Part II of the question naire . Coding Categories for Card No. 1 Column Numbers 1 through 4— Identification. Column No. 1— Nationality 1. Pakistan (not used in this study) 2. United States 3. Brazil Columns No. 2, 3, and 4— Serial numbers of respondents. 482 Column No, 5— Age 0 - Not codeable 1 - Under 25 years 2 - 25 to 29 years 3 - 30 to 34 years 4 - 35 to 39 years 5 - 40 to 44 years 6 - 45 to 49 years 7 - 50 to 54 years 8 - 55 to 59 years 9-60 years and over Column No. 6— Marital status 0 - Not codeable 1 - Married 2 - Not married 3 - Widowed 4 - Separated 5 - Divorced Column No. 7— Number of children 0 - No children 1 - One child 2 - Two children 484 3 - Three children 4 - Four children 5 - Five children 6 - Six children 7 - Seven children 8 - Eight or more children 9 - Not codeable Column No. 8— Father's occupation 0 - Not codeable 1 - Governmental employee 2 * Nongovernmental employee 3 - Self-employed professional, e.g., doctor, lawyer, dentist, accountant, et cetera 4 - Self-employed nonprofessional, e.g., landowner, businessman, et cetera Column No. 9— Number of years in present occupation 0 - Not codeable 1 - 1 to 5 years 2 - 6 to 10 years 3 - 11 to 15 years 4 - 16 to 20 years 485 5 - 21 to 25 years 6-26 years or more Column No. 10— Number of years of education 0 - Not codeable 1-14 years or under 2-15 years 3-16 years 4-17 years 5-18 years 6-19 years 7-20 years or more Column No. 11— Degrees held by respondent 0 - No degree 1 - Bachelor 2 - Master 3 - Doctorate 4 - Not codeable Column No. 12— Field of study of respondent 0 - Not codeable 1 - Liberal arts 2 - Physical sciences 486 3 - Professional, excluding public administra tion, e.g., law, engineering, medicine, et cetera 4 - Public administration 5 - Social science 6 - Other Column No. 13— Specialized training 0 - No special training 1 - Special training in public administration 2 - Special training in fields other than public administration 3 - Not codeable Column No. 14— Religion of respondent 0 - Not codeable 1 - Protestant 2 - Catholic 3 - Jewish 4 - Muslim 5 - Other 487 Coding Categories for Card No. 2 Each of the twenty-four questions in Part II of the questionnaire offered the respondent a choice of responses: (1) in some cases one out of four; (2) in some one out of three; and (3) in some one out of two. Each response had a number assigned to it. Thus, if there were four alternatives, they were numbered one, two, three, and four. The respondent indicated his preference by writing the number of the response he considered appropriate on x the line provided for each question. There were two categories of response to each of these questions. Under one column, the respondent stated what actually happens in each hypothetical situation; and under the other, he stated what he felt should happen. The responses were punched directly according to their original numbers, with uncodeable responses being punched as 0. Tables 1 to 6 indicate which alternative responses to every question were codified as indicative of more position-orientedness and which ones were codified as indicative of less position-orientedness. In those tables, both questions and alternative responses keep the 488 same number as they appear In Appendix A. The first column of card number 2 identifies the nationality of the respondent according to the following code: 1. Not used 2. United States 3. Brazil Columns 2 through 4 identify the serial number of the respondent. Columns 5 and 6 identify the responses to question number 1, and the coding categories are the same as the original numbers assigned to each alternative response. Similarly, columns 7 and 8 are for question number 2, and so on. APPENDIX C TIME PATTERNS QUESTIONNAIRE TIME PATTERNS QUESTIONNAIRE Question Number Background Information 1. Fhat country do you come from? 2. Vhat is your age?_____________ 3. Have you held an administrative position in your career? 1. Yes 2._______ No 4. If you have held an administrative position, how many years have you been in such work? Perception of Time as a Scarce Commodity 5. If you had agreed to meet a person at 11:00 A.M. on business, at what time would you consider him late for his appointment with you? __________ (Please give time to minutes, such as 11:31.) 6. At what time would you decide there was no point in waiting further? __________ (Please give time to minutes, such as 12:31, 1:40.) 490 491 7. If the person were a friend and the purpose of the meeting were social, what would be your feelings about his arriving at 11:30? 1. _____Very angry 2. _____ Somewhat angry 3. _____Displeased, but not angry 4. Unconcerned 5. _____ Pleased 8. T'foat is the length of the official work week for government employees in your country? ____________ (Give exact number of hours, such as 32, 34, 40, etc.) 9. As you look back on your own experience, would you generally say that in work (or school) situations you have felt rushed, with not enough time to get the things done you want to do? 1._______ Very common 2. _____ Somewhat common 3. _____ Not very common 4. _____ Very uncommon 10. What is your feeling about the people you have observed in your government service? Do they seem to be trying to do more than they can fit into the time available or not? 1._______ Very frequently seem to be working at high pace 2. _____ Frequently seem to be working at high pace 3. _____ Infrequently seem to be working at high pace 4. _____ Very infrequently seem to be working at high pace 492 11. As a rough guess, how many hours per week do you think that top level professional people in the government are actually on the job? __________ (Give exact number of hours, such as 50, 54, etc.) 12. How would you compare the amount of time spent on the job by a top level professional person in the government with one in private business? 1. _____ Government official spends considerably more time on job 2. _____ Government official spends slightly more 3. _ _ _ About equal 4* _____ Government official spends slightly less time on job 5* _____ Government official spends considerably less 13. How would you compare the amount of time spent on the job by a clerical person in the government with one in private business? — — . Government employee spends considerably more time on job. 2. __ Government employee spends slightly more 3* — — _ About equal 4. __Government employee spends slightly less time on job, 3* _____ Government employee spends considerably less 14. How often do you receive people without any prior appointment ? 1. __Very frequently 2. Frequently 493 3. Occasionally 4. Not at all 15. How do you feel about receiving visitors who have not made a prior appointment? Highly disturbed 2. ____ Somewhat disturbed 3, . Indifferent - Somewhat pleased 5._______ Very pleased 16. Do you feel that your typical work day proceeds according to an orderly plan? 1. ______ Always 2. — . Quite frequently 3.________ Occasionally _ Not at all 17. Do you feel that your typical work day should proceed according to an orderly plan? 1« . A great deal more than presently 2. ______Somewhat more than presently 3« - Should be about the same — Somewhat less than presently -*•-----A great deal less than presently 494 13. In planning and thinking about your job future, what is the most common future period in which you deal? 1. The next month The next year 3. — The next three years 4. _____ The next ten years 4. The years beyond the next ten 6. Never think about the future. 19. At the time of making a decision, what considera tions guide you the most? (Rank order, please) Present official policy Future implications _ _ Past procedures 20. As you think about your government administration in very broad terms, would you characterize it as one that seems to be most preoccupied with _ the past, the present, ____ the future? 21. What time period do you think a government should be preoccupied with? the past, the present the future 22. If you have worked in your government, what time period do you think your job was most concerned with? the past, the present, ____ the future 495 23. What kind of tims period do you feel you would be most interested in having your job concerned with? the past, ____ the present, ____ the future 24. What percentage of your time spent with people in your office is primarily of a social nature; that is, having refreshments, talking about nonofficial matters, etc? 1. ____ Less than 10 per cent 2. ____ 10 to 25 per cent 3. ____ 26 to 50 per cent 4. ____ More than 50 per cent 25. How common is it for a top level professional in the government to hold other employment as well? 1. ____ Very common 2. ____ Fairly common 3. ____ Neither common nor uncommon 4. ____ Fairly uncommon 5._____ Very uncommon 26. If it is fairly common or very common for top profes sional people to hold dual employment, what percent age of time would you say is generally spent on government business? 1._____ 75 per cent or more on government business 2. 50 to 74 per cent on government business 3. ____ 25 to 49 per cent on government business 4. Less than 25 per cent on government business 496 27. How common is it for a clerical person in the govern ment to hold other employment as well? 1._______ Very common 2. _____ Fairly common 3. _____ Neither common nor uncommon 4. _____ Fairly uncommon 5. _____ Very uncommon 28. If it is fairly common or very common for clerical people to hold dual employment, what percentage of time would you say is generally spent on government business? 1._______ 75 per cent or more on government business 2._______ 50 to 74 per cent on government business 3. _____ 25 to 49 per cent on government business 4. _____ Less than 25 per cent on government business APPENDIX D TIME PATTERNS— CODING CATEGORIES CODING CATEGORIES The data from the questionnaires were punched on IBM cards, one card for each questionnaire. Listed below are the column numbers which corres pond to the twenty-eight items of the questionnaire. As it will be noticed, the column numbers do not follow an uninterrupted order, whereas the items of the question naire run from 1 to 28, consecutively. This is due to the fact that the questionnaire, as presented in Appendix C, is already a rearrangement, and not the original form, of the instrument applied to the respondents. Therefore, interruptions in the sequence of the column numbers correspond to items of the original questionnaire, which, though punched on the IBM card, were irrelevant for the purposes of this study. The majority of items was already precoded (see Appendix C). Therefore, to avoid repetition, only those coding categories that do not appear in the instrument are presented here. 498 499 Column Que s t ion Number Information Number 1 Nationality 1 1. United States 2. Brazil 2-3 Age (reduced to years, as all similar 2 items) 4 Supervision experience 3 5-6 Length of supervision experience 4 17-18 Definition of lateness (reduced to 5 minutes, as all similar items) 19-20 Tolerance of lateness 6 22 Reaction to lateness on social matters 7 48-49 Prescribed working load 8 67 Respondent’s felt sense of urgency 9 68 Respondent's perception of governmental 10 personnel's work pace 56-57 Time inputs of top governmental 11 personnel 58 Relative time inputs, at higher echelons, 12 between public and private personnel 59 Relative time inputs, at lower echelons, 13 between public and private personnel 27 Scheduling habit (appointments) 14 28 Reaction to lack of scheduling 15 (appointments) 500 Column Question Number Information Number 37 Scheduling habit (typical work day) 15 perception 38 Scheduling habit (typical work day) 17 expectations 25 Respondent1s time perspective 18 32-36 Direction of time perspective in 19 governmental decision-making 69 Respondent's perception of his govern- 20 ment's temporal orientations--!. Past, 2. Present, 3. Future (as all similar items) 70 Respondent's expectation about his 21 government's temporal orientations 72 Respondent's perception of his job's 22 temporal orientations 71 Respondent's expectation for his 23 job's temporal orientations 47 Job-unrelated allocation of time on 24 the job 50 Dual employment in government, at 25 higher echelons 61 Time inputs affected by dual employment 26 in government, at higher echelons 62 Dual employment in government, at 27 lower echelons 63 Time inputs affected by dual employment 28 in government, at lower echelons

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A Systematic Method Of Personnel Evaluation: Appraising The Job Performance Of Instructors In Higher Education By The Use Of Decision Theory

PDF

Authority Behavior: An Empirical Approach To Comparative Public Administration

PDF

The Administrative Process And Decision-Making

PDF

Changing Patterns In British Administration In Pakistan And India: A Study In Bureaucratic Attitudes

PDF

Role Conflict Resolution In Complex Social Systems: A Polyarchical Systems Model For Organizational And Social Change

PDF

Community Economic Development And Upward Social Mobility Of Black Students: A Strategy For Social Change

PDF

California State Training Division: A Study In Institution Building

PDF

A Study Of The Role And Function Of The Prison Officer

PDF

Pay Utility In The Public Sector

PDF

Juvenile Delinquency: Linkage Between Maintenance And Precipitating Variables As Part Of A System

PDF

The Image Of The Municipal Service: A Study Of Occupational Values Of Professional Personnel

PDF

The Implementation Of Management By Objectives In Public Sector Organizations

PDF

The Problem Of Highway Location And Contract Policies In The State Of California

PDF

An Analysis Of The Role And Communications Dynamics Of The Assistant-To The City Manager

PDF

Performance Improvement Through Self-Contracting

PDF

Administration Of Land Reform In Iran

PDF

Autonomic and cognitive indices of semantic conditioning and generalization

PDF

Participative Management And Industrial Democracy: Toward A Participative Theory Of Organization

PDF

Alternatives In Research Design: The Measurement Of Goal Conflict

PDF

Multipurpose District Legislation For California: 1958-1963

Asset Metadata

Creator Nascimento, Kleber Tatinge Do (author)

Core Title Change Strategy And Client System: Administrative Reform In Brazil

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Public Administration

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,political science, public administration

Language English

Advisor Sherwood, Frank P. (committee chair), Gable, Richard W. (committee member), McEachern, Alexander W. (committee member), Siegel, Gilbert (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-211264

Unique identifier UC11359989

Identifier 6607078.pdf (filename),usctheses-c18-211264 (legacy record id)

Legacy Identifier 6607078.pdf

Dmrecord 211264

Document Type Dissertation

Rights Nascimento, Kleber Tatinge Do

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA